macroeconomic challenges of scaling up aid to africa...millennium project (2005) has argued for $33...

Macroeconomic Challenges ofScaling Up Aid to AfricaA Checklist for Practitioners

Sanjeev Gupta, Robert Powell, andYongzheng Yang

International Monetary FundWashington, DC

©International Monetary Fund. Not for Redistribution

© 2006 International Monetary Fund

Production: IMF Multimedia Services DivisionCover Design: Julio Prego

Typesetting: Alicia Etchebarne-BourdinCover Photo: Brent Stirton/Getty Images

Cataloging-in-Publication Data

Gupta, Sanjeev.Macroeconomic challenges of scaling up aid to Africa: a checklist for practitioners/

Sanjeev Gupta, Robert Powell, and Yonzheng Yang—Washington, D.C.: InternationalMonetary Fund, [2006].

p. cm.

ISBN 1-58906-505-0Includes bibliographical references.

1. Economic assistance—Africa—Handbooks, manuals, etc. 2. Africa—Economicpolicy—Handbooks, manuals, etc. 3. Monetary policy—Africa—Handbooks, manuals,etc. 4. Debts, External—Africa—Handbooks, manuals, etc. I. Powell, Robert, 1962–II. Yang, Yongzheng.HC800.G87 2006

Disclaimer: This publication should not be reported as representing the viewsor policies of the International Monetary Fund. The views expressed in thiswork are those of the authors and do not necessarily represent those of theIMF, its Executive Board, or its management.

Price: $25.00

Please send orders to:International Monetary Fund, Publication Services

700 19th Street, NW, Washington, DC 20431, U.S.A.Telephone: (202) 623-7430 Telefax: (202) 623-7201

Internet: http://www.imf.org

East London, South Africa: Children play on a swing at a community daycare center for disadvantaged children.


http://www.imf.org

Contents

Glossary of Terms and Abbreviations vii

Preface ix

Acknowledgments x

1. The Macroeconomic Effects of Scaling Up Aid 1

1.1. Meeting the Millennium Development Goals 11.2. Tracking Recent Aid Flows to Africa 21.3. Using Scaling-Up Scenarios to Identify Appropriate Policies 21.4. Tailoring the Scaling-Up Scenario to the Country’s Needs 3

1.4.1. Costing Non-Income-Related MDGs 41.4.2. Projecting the Macro Impact of Various Aid Levels 41.4.3. Targeting a Particular Growth Rate 4

1.5. Estimating Additional Aid and Identifying Appropriate Policies 41.5.1. Quantifying New Aid Flows 51.5.2. Analyzing Key Policy Questions 51.5.3. Using a Standard Macroeconomic Framework 61.5.4. Distinguishing between the Short and Long Term 61.5.5. Assuming a Strengthened Policy Environment 71.5.6. Building Alternative Scenarios 71.5.7. Monitoring and Evaluating the Impact of Aid 7

1.6. The Organization of this Handbook 7

2. Managing the Real Exchange Rate 9

2.1. Absorbing and Spending Aid Flows 92.1.1. Aid Absorption 92.1.2. Aid Spending 102.1.3. Absorption and Spending Policy Options 11

2.2. Adjusting the Real Exchange Rate 122.2.1. Dutch Disease 122.2.2. Supply and Demand Effects 132.2.3. The Impact on Exchange Rates and Growth 14

3. Using Monetary Policy for Sterilization and Inflation Management 16

3.1. Policy Options for Sterilization 16

iii


Underline

Underline

Underline

Underline

Underline

Underline

Underline

Underline

Underline

Underline

Underline

Underline

Underline

Underline

Underline

Underline

Underline

Underline

Underline

Underline

Underline

Underline

Underline

Underline

Underline

Underline

Underline

Underline

Underline

Underline

Underline

3.1.1. Domestic Sterilization Measures 163.1.2. Foreign Sterilization 173.1.3. Domestic Versus Foreign Sterilization 17

3.2. Managing Inflation 18

4. Mobilizing Adequate Domestic Revenues 19

4.1. Preventing Aid Dependency 194.2. Protecting Revenues 20

4.2.1. Loans Versus Grants 214.2.2. Tax Treatment of Foreign Aid 214.2.3. Revenue Assumptions for Scaling-Up Scenarios 22

5. Projecting the Impact of Increased Aid on Economic Growth 23

5.1. The Relationship between Aid and Growth 235.1.1. The Impact of Different Types of Aid 245.1.2. The Growth Effects of Scaling Up 24

5.2. Accounting for Diminishing Returns and Limits to Absorptive Capacity 25

5.2.1. Convergence Parameters 255.2.2. Diminishing Returns to Aid 255.2.3. Postconflict Situations 26

5.3. Safeguarding Private Savings and Investment 265.4. Raising Spending as a Share of GDP 275.5. Confirming the Positive Impact of Aid 28

6. Meeting Other Fiscal Challenges 30

6.1. Developing an Exit Strategy 306.1.1. The Scaling Down of Aid 306.1.2. Private Economic Activity 316.1.3. Real Spending Levels 31

6.2. Properly Estimating Current Spending 316.3. Targeting the Poor 326.4. Containing Unproductive Spending 326.5. Balancing Poverty Reduction and Growth 326.6. Minimizing Bottlenecks and Improving Coordination 33

6.6.1. Bringing NGO Activities into the Budget 336.6.2. Government at Lower Levels 346.6.3. Existing Private Sector Capacity 34

7. Strengthening Governance 35

7.1. Reducing Corruption 357.1.1. The Impact of Corruption on Economic Performance 367.1.2. Successful Anticorruption Strategies 367.1.3. Breaking the Cycle of Poor Governance 37

7.2. Improving Public Expenditure Management Systems 37

8. Maintaining Debt Sustainability 38

8.1. Assessing External Debt Dynamics 38

iv MACROECONOMIC CHALLENGES OF SCALING UP AID TO AFRICA


Underline

Underline

Underline

Underline

Underline

Underline

Underline

Underline

Underline

Underline

Underline

Underline

Underline

Underline

Underline

Underline

Underline

Underline

Underline

Underline

Underline

Underline

Underline

Underline

Underline

Underline

Underline

Underline

Underline

Underline

Underline

Underline

Underline

Underline

Underline

Underline

Underline

Underline

Underline

Underline

Underline

Underline

8.2. Gauging Fiscal Debt Sustainability 398.3. Ensuring Debt Sustainability 408.4. Strengthening the Debt-Management System 41

9. Summary: Five Primary Guidelines 43

9.1. Minimize the Risks of Dutch Disease 439.2. Seek to Enhance Growth in the Short to Medium Term 449.3. Promote Good Governance and Reduce Corruption 449.4. Prepare an Exit Strategy 459.5. Regularly Reassess the Appropriate Policy Mix 45

Appendix 1. The Relationship between Aid Flows and Exchange Rates in Sub-Saharan Africa 46

Appendix 2. The Macroeconomics of Aid 49

References 59

Box

1. Absorption of Financing for HIV/AIDS Programs 10

Figures

1. Impact of Scaling Up on External Debt 392. Impact of Scaling Up on Fiscal Debt 40

Tables

1. Patterns of Aid Inflows in Five African Countries 32. Possible Combinations of Absorption and Spending in

Response to a Scaling Up of Aid 123. Mitigating Aid Volatility 20

Contents v


Underline

Underline

Underline

Underline

Underline

Underline

Underline

Underline

Underline

Underline

Underline

Underline

Underline

Underline

Underline

Underline

Underline

Underline

This page intentionally left blank


Glossary of Terms and Abbreviations

CPI Consumer price indexCPIA Country Policy and Institutional Assessment (World Bank)DfID Department for International Development (U.K.)DSA Debt sustainability analysisDTIS Diagnostic Trade Integration StudyG-8 Group of Eight industrial countriesHIPC Highly indebted poor countryIF Integrated FrameworkIMF International Monetary FundITC International Trade CenterLIC Low-income countryMCA Millennium Challenge Account (U.S.)MDGs Millennium Development GoalsMIC Middle-income countryMTEF Medium-Term Expenditure FrameworkNGO Nongovernmental organizationNPV Net present valueODA Official development assistanceOECD Organization for Economic Cooperation and DevelopmentPEM Public expenditure managementPEPFAR President’s Emergency Program for AIDS Relief (U.S.)PER Public Expenditure Review (World Bank)PRSP Poverty Reduction Strategy PaperSSA Sub-Saharan AfricaUN United NationsUNCTAD United Nations Conference on Trade and DevelopmentUNDP United Nations Development ProgramVAT Value-added taxWHO World Health OrganizationWTO World Trade Organization

vii


This page intentionally left blank


Preface

In 2005, major donor countries of the Organization for EconomicCooperation and Development indicated their intention to increase signifi-cantly the amount of external assistance they provide to developing countriesin the next decade. The sharpest increase is likely to be in Africa, where a scal-ing up of external assistance, including comprehensive debt relief, will aim tosupport individual countries’ efforts to achieve the Millennium DevelopmentGoals (MDGs). This handbook was prepared in the Policy Wing of the IMF’sAfrican Department as a practical guide for assessing the macroeconomicimplications and challenges associated with a significant scaling up of aid toAfrican countries. Its purpose is to provide a resource for policymakers, prac-ticing economists in African countries, and staff of international financialinstitutions and donor agencies who participate in the preparation ofmedium-term strategies for individual African countries, including in the con-text of Poverty Reduction Strategy Papers (PRSPs). A version of this was pub-lished in September 2005 as an IMF Working Paper (WP/05/179).

When aid flows increase, a country has to choose how much to absorb andhow much to spend. When aid is fully absorbed, a real appreciation of theexchange rate is likely, at least in the short run. In a number of countries, aidsurges have been associated with real exchange rate depreciation, suggestingthat supply effects have more than offset the impact of higher domesticdemand on the nontradable sectors or, alternatively, that the aid has not beenfully absorbed. Developing a good understanding of how a country will spendadditional aid, as well as the appropriate policy response of the central bank,is therefore critical to projecting the likely macroeconomic impact. Thishandbook helps policymakers and economists develop various “scaling-up sce-narios” to assess such contingencies. Scaling-up scenarios are not forecasts.They are tools to help countries identify important policy issues involved inusing higher aid flows effectively and to help both donors and recipients pro-vide a regularly updated vision for strengthening the potential impact of aidon growth, especially through policies that strengthen economic governance.Governments that receive higher aid flows also face the challenge of what todo if and when such aid flows are interrupted or reduced, and hence theyneed an “exit strategy” as part of the scaling-up scenario.

Benedicte Vibe ChristensenDeputy Director

African Department

ix


Acknowledgments

The authors would like to thank Kevin Carey, Jakob Christensen, MarkusHaacker, and Smita Wagh for their contributions, and David Bevan for pro-viding extensive comments on earlier drafts. Very helpful comments were alsoreceived from IMF colleagues in the African, Fiscal Affairs, PolicyDevelopment and Review, and Research Departments, as well as the WorldBank and the United Kingdom’s Department for International Development(DfID). The authors are, however, solely responsible for any remaining errors.Linda Griffin Kean of the External Relations Department of the IMF editedand coordinated the publication of the paper, and Alicia Etchebarne-Bourdin, also of the External Relations Department, typeset the book.

This publication should not be reported as representing the views or poli-cies of the International Monetary Fund. The views expressed in this work arethose of the authors and do not necessarily represent those of the IMF, itsExecutive Board, or its management.

x


1The Macroeconomic Effects of

Scaling Up Aid

This handbook provides a checklist of the macroeconomic challengesthat low-income countries are likely to face if they begin to receive signifi-cantly higher official development assistance (ODA) than in the recent past.The checklist, which is derived from a survey of the economic literature, is atool for developing illustrative macroeconomic scenarios for individual coun-tries in response to a scaling up of aid flows. For example, one scaling-up sce-nario might involve a doubling of official resource transfers as a share of arecipient country’s GDP, with higher aid flows being sustained for a decadeor more.

1.1. Meeting the Millennium Development Goals

Such scenarios are most likely to be developed in the context of a coun-try’s efforts to achieve the Millennium Development Goals (MDGs) with thesupport of the international donor community.

Estimates of the costs of reaching the MDGs vary widely. The 2005Group of Eight (G-8) Gleneagles communiqué (Group of Eight, 2005) callsfor raising annual aid flows to Africa by $25 billion by 2010, while the UNMillennium Project (2005) has argued for $33 billion of additional annualresources to achieve the MDGs in Africa. Consistent with the G-8 proposals,the World Bank and IMF (2005) argue that a conservative estimate of theadditional ODA that Africa could use effectively in both infrastructure andhuman development ranges from $14 to $18 billion per year during2006–08, and rises to $24–$28 billion by 2015. ODA (including debt relief)to sub-Saharan Africa (SSA) averaged about $17 billion per year during2000–03.

1

Scaling-up scenarios mayhelp guide a country’s efforts

to achieve the MDGs.


1.2. Tracking Recent Aid Flows to Africa

Data from the World Bank (2005) suggest that net official transfers toindividual African countries have varied considerably over the past quartercentury (see Appendix Table A.2.1). Of the 44 countries covered by the IMF’sAfrican Department, 14 have received net official transfers of 5 percent ofGDP or less, on average, between 1980 and 2003. These are generally thewealthier countries in terms of GDP per capita and some are oil producers.Twenty-four countries received between 6 and 16 percent of GDP on average,while a group of six countries, mainly small economies, received andabsorbed transfers in excess of 20 percent of GDP on a sustained basis. Manycountries have also received large aid inflows following the cessation of conflict.1

A few countries in Africa have experienced modest surges of aid in therecent past (Table 1).2 All five countries listed in the table received debt reliefover the period, which, in turn, permitted some of them to clear their exter-nal arrears and increase their net aid inflows. In gauging the impact of an aidsurge, private inflows (for example, foreign direct investment) need to beconsidered along with the public inflows. If, for example, a surge in aid is off-set by a corresponding fall in private inflows (as for Mozambique), the chal-lenge of macro management is considerably different.

1.3. Using Scaling-Up Scenarios to Identify Appropriate Policies

The goal of preparing medium-term scaling-up scenarios is to identifysome of the key measures and policies that can help a country absorb a higherlevel of aid and ensure that it uses the aid efficiently. Scaling-up scenarios arenot meant to be unconditional forecasts of the actual impact of higher aidflows. Instead, they are designed to illustrate the potential impact on a coun-try of a sustained increase in external aid if the country also implements eco-nomic policies that allow it to use and absorb the assumed aid flows withoutdamaging or destabilizing the macroeconomic environment. Scaling-up sce-narios may help countries avoid the type of absorption problems that maymake donors less likely to offer higher aid and may make recipient govern-ments less likely to spend it on a sustained basis—problems such as risinginflation, severe skill shortages or other bottlenecks, or a serious loss of inter-national competitiveness. However, scaling-up scenarios are not meant to

2 MACROECONOMIC CHALLENGES OF SCALING UP AID TO AFRICA

1See Collier and Hoeffler (2004). The ability of postconflict countries to absorb aid is some-what different from that of countries that have a reform program supported by the World Bankor the IMF. The challenges facing postconflict countries are discussed in Section 5.2.3.

2Net aid flows are defined as disbursements of grants and loans plus debt relief, net of amorti-zation, interest payments, and the change in external arrears. This measure captures the actualinflows of foreign exchange and, hence, the scale of the macroeconomic challenge faced by therecipient.


identify the point at which an economy will be no longer able to absorbhigher aid flows effectively—that is, the point when the challenges will start tooutweigh the benefits. That point is difficult to predict.

Medium-term scaling-up scenarios can be dynamic tools, because theycan be updated regularly to reflect the changing medium-term visions ofboth donors and recipient countries (see Section 1.5.7 on monitoring andevaluation).

1.4. Tailoring the Scaling-Up Scenario to the Country’s Needs

There are three basic approaches to preparing scaling-up scenarios. Theprecise form of the scenario should reflect the goals of the country and theavailability of other information. Some countries may choose to prepare ascaling-up scenario for inclusion in their PRSP.

The Macroeconomic Effects of Scaling Up Aid 3

Scaling-up scenarios can identify some key

measures and policies tohelp ensure that

countries absorb a higherlevel of aid and use it

efficiently.

Table 1. Patterns of Aid Inflows in Five African Countries(In percent of GDP)

1998 1999 2000 2001 2002 2003

Ethiopia1

Net aid inflows 4.7 6.0 8.8 16.1 15.0Gross aid inflows 11.7 8.8 24.3 18.1 17.5Net private inflows 6.6 8.1 6.8 5.7 7.7

GhanaNet aid inflows 3.2 2.8 –0.3 10.6 2.6 7.1Gross aid inflows 8.7 7.5 8.8 14.9 6.1 9.5

of which program aid 1.8 1.9 3.8 5.6 2.6 5.1Net private inflows 6.0 6.3 11.2 13.0 12.0 13.7

MozambiqueNet aid inflows 11.6 11.4 20.4 15.4 16.4 15.0Gross aid inflows 13.4 13.4 20.0 16.7 18.5 17.4


Tanzania1

Net aid inflows 4.6 6.6 7.5 7.9 6.6 7.6Gross aid inflows 13.3 12.7 12.8 12.5 10.5 10.5


Uganda1

Net aid inflows 6.5 6.8 15.5 10.5 9.9Gross aid inflows 10.0 11.1 16.3 14.6 12.4

of which program aid 2.5 3.6 8.3 7.6 6.7Net private inflows 3.0 3.2 2.8 3.2 3.3

Source: IMF (2005d).Note: Figures in bold represent periods of aid surges.1In Ethiopia, Tanzania, and Uganda, the fiscal year begins in July. Hence, for example, 1999

data covers July 1998 through July 1999.


1.4.1. Costing Non-Income-Related MDGs

This approach assesses the macroeconomic implications of a fiscal scenariothat is based on an explicit costing of achieving those MDGs that do not focuson income levels (for example, those related to health, education, and wateraccess). The costing exercise, which is typically carried out with assistance fromdevelopment partners such as the World Bank and the United Nations (UN),provides a judgment about the resources required in each sector. It also mayillustrate some trade-offs among policies, resources, and macroeconomic out-comes, and may identify supply bottlenecks that need to be addressed.

1.4.2. Projecting the Macro Impact of Various Aid Levels

A second approach is to assess the macroeconomic impact of a significantbut arbitrary increase of external assistance (for example, a doubling of aid inproportion to GDP). Here, the higher level of resources is assumed, but it isnot grounded in an explicit costing of the MDGs. The scaling-up scenarioindicates how these resources might be used in a fiscal projection (includingpossible trade-offs among competing sectors or MDGs) and the potentialimpact of the higher spending on key macroeconomic indicators. Thisapproach may be appropriate when an explicit MDG costing is not availableor when the implied additional resources are judged to be too large for acountry to absorb without harmful macroeconomic implications. Even in theabsence of a comprehensive costing of the MDGs, preparing broad-brush scal-ing-up estimates serves to put important issues on the table for discussionamong country authorities and donors.

1.4.3. Targeting a Particular Growth Rate

A third approach is to assess the implications of a specific target growthrate in an environment of scaling up. Target growth rates are given with a viewto achieving the income-poverty MDG; higher external resources are assumedto be available; and the analysis suggests the kind of improvements in pro-ductivity and/or other policies that might be required to meet the macroeco-nomic goals for a given increase in aid.

1.5. Estimating Additional Aid and Identifying Appropriate Policies

The amount of additional aid and the macroeconomic and sectoral poli-cies required to achieve the MDGs will vary from country to country. To deter-mine the appropriate assumptions for developing the medium-termscaling-up scenario, countries should seek assistance and advice from devel-opment partners, particularly the World Bank. For some countries, the WorldBank, the UN, or other donors may have already prepared detailed sectoral



assessments of the estimated costs and the associated sectoral policies neces-sary to achieve individual MDGs.3

When such costing analysis is available, it forms an important foundationfor the associated macroeconomic projections of scaling up. As noted, how-ever, when a fully articulated MDG costing exercise is not available, it may benecessary to use more broad-brush assumptions as the basis for scaling-up pro-jections. Moreover, if the projected aid requirements are considered too unre-alistic, it may be appropriate to prepare a less ambitious scaling-up scenariothat assumes more limited external resources and therefore illustrates thepotential trade-offs among competing policy goals.

1.5.1. Quantifying New Aid Flows

Under all three approaches, a number of fundamental assumptions orjudgments must be made about the level of increased aid. Key among theseare the following:

• the expected level of additional external assistance

• the time period over which the increased aid will be available

• how quickly assistance will fall back to “normal” levels

• the sectors in which additional resources will be spent (for example,health, education, roads, infrastructure)

• how long a country’s individual sectoral budgets will be increased

• whether the near-term (for example, three-year) financing assumptionsare sufficiently realistic to be incorporated into the country’s medium-term expenditure framework (MTEF)

• whether additional assistance will take the form of project finance orbudget support, will be provided as grants or loans, or will be tied aid

• the balance between current and capital expenditures, and the share ofadditional public spending likely to be earmarked for imports

• the recurrent cost implications of public investment spending.

1.5.2. Analyzing Key Policy Questions

The scaling-up scenario must analyze the following key policy questions:

• How will private investment and savings be affected by higher govern-ment spending and other policies to be undertaken?


Estimating the costs ofachieving each MDG

highlights the potentialtrade-offs among

policy goals.

3For example, for Ethiopia, the World Bank prepared scaling-up scenarios in the context of the2003 Public Expenditure Review (PER). More recently, the Bank has developed a standard(recursive) dynamic general equilibrium model with an additional MDG module that links spe-cific MDG-related interventions to MDG achievement (see Lofgren and Diaz-Bonilla, 2005). Fora full description of the IMF’s scaling-up scenario for Ethiopia, see IMF (2005b and 2005c) andAndrews, Erasmus, and Powell (2005).


• How will real GDP growth and productivity be affected by higher expen-ditures (including private investment), and how quickly will theseresults be seen?

• How relevant is governance for safeguarding public resources andensuring their efficient use?

• What will be the impact of higher aid on the real exchange rate andtrade volumes, and over what time period?

• What are the implications of higher resource flows for monetary policymanagement?

• How will domestic revenue be affected by higher aid flows? Does thecomposition of aid (grants or loans) affect revenue performance?

• What will be the impact of higher aid inflows on debt sustainability?

1.5.3. Using a Standard Macroeconomic Framework

Using a scaling-up scenario to assess the macroeconomic impact ofincreased aid is, in principle, no different from making any other macroeco-nomic projection. All the basic identities that form the foundation of theIMF’s financial programming framework continue to hold:

• The budget deficit must be fully financed from either external ordomestic sources.

• Projections for reserve money and broad money growth should be con-sistent with the output and inflation forecasts and based on realisticassumptions about the velocity of circulation and the money multiplier.

• Real GDP growth and trade projections should be consistent with theassumed path for both public and private investment, the real exchangerate, private credit growth, and expected developments in productivity,including those resulting from scaling up.

• Debt sustainability (both external and domestic) needs to be reassessedand maintained through a prudent debt-management strategy.

• The scaling-up scenario must be accompanied by a set of country-spe-cific policy assumptions that provides some validation for the embodiedeconomic assumptions.

1.5.4. Distinguishing between the Short and Long Term

An important characteristic of the scaling-up scenario is its long timehorizon, which contrasts with the shorter period covered by a typical financialprogramming exercise. The longer time horizon allows for assessment of boththe sustainability of debt levels under the various MDG scenarios and theimpact of higher spending on growth and on non-income MDGs. It is impor-tant to decompose budget expenditures into 1) those that enhance growth inthe short to medium term and 2) those that boost growth in the long run.


The basic identities thatform the foundation of theIMF’s financial program-ming framework continueto hold.


1.5.5. Assuming a Strengthened Policy Environment

The assumptions made about the policy environment (including gover-nance) are critical to assessing the likely impact of higher aid flows on macro-economic performance. For example, scenarios that incorporate the adop-tion of stronger policies illustrate how constraints on the absorption of higheraid flows can be alleviated. Moreover, it is critical to assume the implementa-tion of a policy framework that encourages increased private sector saving,and hence a “crowding in” of private sector investment, along with theassumed higher levels of public investment.

1.5.6. Building Alternative Scenarios

The fragility of many of the assumptions and the long time period underconsideration argue that a range of scenarios be analyzed. While discussionsof scaling up are likely to focus primarily on a single, best-case scenario, alter-native scenarios are helpful to illustrate the implications of a range of possi-ble policy environments and outcomes. For example, alternative scenarioscan be used to “accommodate” fundamental differences between the policiesand assumptions that donors, the IMF, and other international financial insti-tutions consider plausible and those the country authorities seek to adopt—for example, different aid levels, exchange rate or growth targets, or budgetexpenditures.

1.5.7. Monitoring and Evaluating the Impact of Aid

Scaling-up scenarios must be regularly updated and revised. In thisregard, it is necessary to continuously monitor and evaluate the impact ofhigher aid flows on important macroeconomic variables, including for exam-ple, wages, prices, and export volumes. It is also necessary to ascertainwhether the assumptions underlying the scenarios are holding true, and, ifnot, what alternative policies or assumptions could be considered. For exam-ple, if the original assumptions about the effects of aid were unduly pes-simistic or optimistic, new scenarios could be developed to accommodate anyunexpected outcomes associated with the original scenario.

1.6. The Organization of this Handbook

This handbook offers a checklist for preparing scaling-up scenarios, usinga survey of the recent empirical literature to identify the key policy issues tobe considered and the primary assumptions and judgments that shouldunderlie such scenarios. Section 2 looks at issues related to the real exchangerate, including the impact of aid on the real exchange rate and the impact ofa rising real exchange rate on exports and competitiveness. Section 3 consid-ers issues in monetary management and projecting inflation. Section 4 dis-cusses the potential impact of a scaling up of aid on revenue mobilization, andhighlights the importance of maintaining domestic revenue levels throughout


Alternative scenarios canillustrate the implicationsof a range of policies and

results over the longertime horizon involved in a

scaling-up scenario.


the scaling-up period. Section 5 examines the overall and the particulareffects of higher public expenditure on growth, emphasizing that the impactof aid depends on how it is spent and the policy environment into which it isdisbursed. Section 6 considers other fiscal issues, including the importance ofensuring sufficient current spending, containing unproductive expenditures,targeting the poor, and monitoring sectoral bottlenecks. Section 7 stresses theimportance of good governance for enhancing growth, improving aid absorp-tion, and building a framework for effective public expenditure management.Section 8 examines issues related to external and fiscal debt sustainability.Section 9 summarizes the five primary guidelines included in this study.Appendix 1 reviews some of the recent literature on which this handbook isbased. Appendix 2 summarizes the quantitative relationships involved.



2Managing the Real Exchange Rate

2.1. Absorbing and Spending Aid Flows

A key issue in assessing the macroeconomic implications of scaling upofficial resource transfers to Africa is the potential impact on the realexchange rate, exports, and competitiveness. Standard analysis suggests thatforeign aid flows augment domestic resources and therefore leave the econ-omy, as a whole, better off. In practice, however, the macroeconomic impactof aid depends both on how a country spends the resources and on its policyresponse. The interaction of fiscal policy with monetary and exchange ratemanagement is key. To highlight this interaction, IMF (2005d) discusses tworelated but distinct concepts: absorbing aid and spending aid.

2.1.1. Aid Absorption

Aid absorption is defined as the extent to which a country’s nonaid cur-rent account deficit widens in response to an increase in aid inflows.4 Thiscaptures the quantity of net imports financed by the increased aid and repre-sents the additional transfer of real resources enabled by the aid. Absorptioncaptures both the direct and the indirect increases in imports financedthrough aid, that is, the government’s direct purchases of imports as well assecond-round increases in net imports resulting from aid-driven increases ingovernment or private expenditures. For a given fiscal policy, absorption iscontrolled by the central bank through two mechanisms: 1) how much of theforeign exchange associated with aid it chooses to sell, and 2) its interest ratepolicy, which influences the demand for private imports through aggregate

9

Aid absorption is the extentto which a country’s nonaid

current account deficitwidens in response to

increased aid inflows. Itreflects the increase in net

imports resulting fromincreased aid.

4This definition of aid absorption in IMF (2005d) differs from that of domestic absorption (thesum of private consumption and investment, and government expenditure). The nonaid currentaccount balance is the current account balance excluding official grants and interest on externalpublic debt, whereas the nonaid capital account balance is the capital account net of aid-relatedcapital flows, such as loan disbursements and amortization.


demand. The aid is not absorbed, however, if central bank sales of foreignexchange are matched by private accumulation of foreign assets. Box 1 dis-cusses the absorption of aid for HIV/AID programs.

2.1.2. Aid Spending

Aid spending is defined as the widening in the government fiscal deficit(net of aid) that accompanies an increase in aid.5 Spending captures theextent to which the government uses aid to finance an increase in expendi-tures or a reduction in taxation. Even if the aid comes tied to particularexpenditures, governments can choose whether or not to increase the overallfiscal deficit as aid increases. Aid-related expenditure increases can be forimports or for domestically produced goods and services.


Aid spending is defined asthe widening in the gov-ernment fiscal deficit thataccompanies an increasein aid. It indicates theextent to which the gov-ernment uses aid tofinance either an increasein expenditures or a reduc-tion in taxation.

Box 1. Absorption of Financing for HIV/AIDS Programs1

External grants have been the dominant source of funding for HIV/AIDS–relatedexpenditures in low- and middle-income countries. In several countries, financingneeds for HIV/AIDS programs exceed total public health expenditure, and theycould rise to up to 10 percent of GDP for some low-income countries. Thus,HIV/AIDS programs can present challenges similar to those associated with thetype of significant scaling up of aid discussed in this handbook. Absorption ofgrants for HIV/AIDS depends on the composition of spending. Much of theHIV/AIDS–related spending finances antiretroviral drugs, which are imported inalmost all low- and middle-income countries. In the countries with highHIV/AIDS prevalence rates (and thus disproportionally high treatment costs),much of the external financing is likely to be spent on imports, thus mitigatingthe macroeconomic implications of high aid inflows. However, other componentsof HIV/AIDS programs, such as prevention and orphan support, largely take theform of domestic spending on nontraded goods and services.

HIV/AIDS programs also have implications for economic growth and the gov-ernment’s fiscal balance. Programs reduce both the human losses from the dis-ease and the number of new infections. To the extent that productive capacity ispreserved, this has a positive effect on government revenues. At the same time,a successful prevention program means that the demand for HIV/AIDS–relatedservices will eventually decline. Masha (2004), for example, estimates that theseindirect savings will amount to at least 15 percent of the annual costs ofBotswana’s National Strategic Program on HIV/AIDS.

1By Markus Haacker and draws on Haacker (2004a, 2005).

5The deficit net of aid is equal to total expenditures (G) less domestic revenue (T) and isfinanced by a combination of net aid and domestic financing: G – T = Nonaid fiscal deficit = Netaid + Domestic financing. A few countries may also supplement their resources with access tononconcessional borrowing.


2.1.3. Absorption and Spending Policy Options

Absorption and spending are policy choices. If the government spendsaid resources directly on imports or if the aid is in kind (for example, grainor drugs), spending and absorption are equivalent, and there is no directimpact on such macroeconomic variables as the exchange rate, the pricelevel, or the interest rate. But if a country receives foreign exchange resourcesand the government immediately sells them to the central bank, then the gov-ernment must decide how much of the local currency counterpart to spenddomestically, while the central bank must decide how much of the aid-relatedforeign exchange to sell on the market. In general, therefore, spending islikely to differ from absorption.

Four basic combinations of absorption and spending are possible inresponse to a scaling up of aid (Table 2). Each one has different macroeco-nomic implications.

• Aid is absorbed and spent. This is the situation assumed in most scaling-up scenarios. The government spends the aid increment, and the cen-tral bank sells the foreign exchange, which is absorbed by the economythrough a widening of the current account deficit. The fiscal deficit islarger but is financed through higher aid.

• Aid is neither absorbed nor spent. The authorities could choose torespond to aid inflows by simply building international reserves. Thismight be an appropriate short-run strategy if aid inflows are volatile orif the country’s international reserves are too low. In this scenario, gov-ernment expenditures are not increased, and taxes are not lowered.Hence, there is no expansionary impact on aggregate demand and nopressure on either the exchange rate or the price level.

• Aid is absorbed but not spent. Increased aid inflows can be used toreduce inflation. This might be appropriate if inflation levels are exces-sively high. The authorities choose to sell the foreign exchange associ-ated with increased aid inflows to sterilize the monetary impact ofdomestically financed fiscal deficits. The result would typically be slowermonetary growth, an appreciated nominal exchange rate, and lowerinflation. It may also allow for lower domestic debt and interest rates.

• Aid is spent but not absorbed. A final possibility is that the fiscal deficit,net of aid, increases with the jump in aid, but the authorities do not sellthe foreign exchange required to finance additional net imports. Themacroeconomic effects of this fiscal expansion are similar to increasinggovernment expenditures in the absence of aid, except that interna-tional reserves are higher. The increased deficits inject money into theeconomy, and inflation increases.

The composition and quality of spending also affect the impact ongrowth. In assessing the overall macroeconomic impact of aid flows, there-fore, it is important to distinguish between different types of aid (project orprogram), the types of expenditures (capital or current) the aid finances, and

Managing the Real Exchange Rate 11

If the government spendsaid resources directly on

imports or if the aid is inkind, then spending and

absorption are equivalent.When a country receives

foreign exchangeresources, then spending isdetermined by the govern-

ment and absorption isdetermined by the

central bank.


the efficiency with which the aid is used. On the other hand, a completeabsorption of aid through an equivalent increase in imports is unlikely toboost growth directly in the short term, although it may do so throughspillover effects over time. These issues are addressed in Sections 4 and 5.

2.2. Adjusting the Real Exchange Rate

In a donor-financed scaling-up scenario, the assumption is typically thatmost (though not necessarily all) aid is both absorbed and spent. In this case,some real exchange rate adjustment may be necessary and, indeed, appropri-ate in response to a sustained higher level of aid because of the effects on therelative demand for imports and for domestically produced tradables andnontradables.

2.2.1. Dutch Disease

Increased aid boosts demand for both imports and domestically pro-duced nontradable goods, including public services such as health care andeducation. Bevan (2005) notes that the public sector is typically assumed tohave a higher propensity to consume domestically produced goods and ser-vices than the private sector. Thus, the domestic component of demand willlikely be higher if the aid finances higher public expenditures than if itfinances tax cuts, transfers to the private sector, or lower domestic borrowing.A country can import goods directly from the world market, but only domes-tic producers can supply nontradables.

Unless there is considerable excess supply in the economy, the prices ofnontradables must become higher than the prices of tradables (that is, the realexchange rate must rise) in order to encourage resources, including labor, toswitch from the production of exportables to the production of nontradedgoods. As the real exchange rate appreciates, the tradable goods sector contractscompared with the nontradable sector—that is the so-called Dutch disease.


“Dutch disease” describesa scenario in which currency appreciationmakes tradable goods less competitive and leadsto an increase in imports.The result is a shift ofresources away from theproduction of tradablegoods and toward nontradables.

Table 2. Possible Combinations of Absorption and Spending in Response to a Scaling Up of Aid

Aid is absorbed and spent. Aid is neither absorbed nor spent.The government spends the aid. Government expenditures are notThe central bank sells the foreign increased.

exchange. Taxes are not lowered.The current account deficit widens. International reserves are built up.

Aid is absorbed but not spent. Aid is spent but not absorbed.Government expenditures are not The fiscal deficit widens (expenditures

increased. are increased).The central bank sells the foreign The central bank does not sell foreign

exchange. exchange.Monetary growth is slowed; nominal International reserves are built up.

exchange rates appreciate; inflation Inflation increases.is lowered.


Dutch disease effects are likely to be stronger when trade is morerestricted, when production is at full capacity, and when the ability of con-sumers to switch between domestic and imported goods in response to rela-tive price changes is more limited as a result. Increased trade liberalizationcan facilitate aid absorption without leading to Dutch disease effects and istherefore one policy option available. Although Nkusu (2004) stresses that afailure to take sufficient account of idle capacity may lead to excessive con-cern about Dutch disease effects, unemployed capital and labor are relevantonly if they can be brought into productive use in response to higher domes-tic demand. Hence, if critical inputs are in short supply (for example, spe-cialized labor) and cannot be replaced by resources that are in abundantsupply, full capacity can coexist with a generalized unemployment of factors.The mechanism for real exchange rate appreciation varies depending on theexchange rate regime. In a pure float, the central bank sells the foreignexchange associated with the aid, causing a nominal (and real) exchange rateappreciation. In a fixed exchange rate, a period of inflation raises the realexchange rate, with the central bank accommodating higher governmentexpenditures. The increase in aggregate demand and the real appreciationcause an increase in net import demand, forcing the central bank to sell for-eign exchange to defend the fixed nominal exchange rate.

2.2.2. Supply and Demand Effects

The macroeconomic impact of aid is likely to depend on how the aid isused. If aid is used to boost supply capacity, the macroeconomic conse-quences will likely be mitigated. On the other hand, if aid finances social sec-tor spending, the macroeconomic consequences will likely be exacerbated.The interaction of demand and supply effects may cause the real exchangerate to “overshoot” its long-run value. This might mean a real appreciation inthe short run, followed by a real depreciation in the medium term. The costsof such real exchange rate volatility will be high if domestic firms facehigh adjustment costs and if domestic financial markets are relatively under-developed. In these circumstances, exporting firms may run down their cap-ital, lay off skilled workers, or even close down, even though the sector’slong-term prospects are favorable. Aid can directly boost supply capacity. Forinstance, in the Adam and Bevan (2003, 2004) model, aid is used to enhancethe supply response of nontraded goods, moderating the relative priceadjustment. Spending aid on infrastructure provides an instrument forimproving the supply response in their model because of the range and scaleof efficiencies that such spending can bring. Bevan (2005) suggests thatthere may be a case for giving a higher priority to scaled-up infrastructureinvestment than to social sector spending because it will yield a better supplyresponse and will offset some of the adverse macroeconomic consequencesof scaled-up aid. If the government does give a higher priority to improvingsocial indicators in the near term, it may be more effective to fine-tune exist-ing social spending than to allocate new aid flows to the social sector. This isdiscussed further in Section 6.


Using aid to boost capacity can mitigate themacroeconomic effects of

aid inflows; using aid for social spending can

exacerbate them.


2.2.3. The Impact on Exchange Rates and Growth

Once appropriate consideration is taken of the supply-side impact of aidflows, there is no clear presumption as to whether the real exchange rate willappreciate or depreciate over the medium term or whether the tradable sec-tor will expand or contract. This is essentially an empirical issue, on whichindividual country circumstances are likely to differ. (See Appendix 1 for asurvey on the literature about the relationships among aid, exchange rates,and economic growth, and Appendix Table A.2.2 for a summary of attemptsto measure the relationship between aid and the real exchange rate in sub-Saharan Africa.)

The evidence suggests that real exchange rate effects on export growthcan be significant in SSA. Region-specific studies find real exchange ratechanges to be a significant determinant of the share of exports in GDP. Forexample, Rajan and Subramanian (2005a) argue that in countries that receivemore aid, export-oriented, labor-intensive industries grow more slowly thanother industries, suggesting that aid does create Dutch disease. In cases wheresuch negative effects are evident, it is important to ensure that the benefits ofaid to the poor are greater than the costs and that the impact of exchange rateinstability on exports is also considered. To the extent that higher aid flowsalleviate supply bottlenecks, they can offset the effect of an exchange rateappreciation on export growth. Accelerations in economic growth rates areassociated with real depreciation, suggesting that a large real appreciationassociated with scaling up could have long-term growth costs. Several studieshave built on the analysis in Hausmann, Pritchett, and Rodrik (2004) ofjumps in countries’ medium-term growth trends, which they label growthaccelerations and which they find to be strongly correlated with real exchangerate depreciation. This finding has been confirmed for SSA in IMF (2005f),which finds that almost all the sustained growth cases in SSA avoided overval-uation during the growth period. The study also notes the close link betweenavoidance of exchange rate misalignment and macroeconomic stability, rein-forcing the case for aid inflows to be accompanied by prudent macroeco-nomic management. In particular, overvaluation of the exchange ratedampens growth, while there is no statistically significant relationshipbetween undervaluation and growth (Razin and Collins, 1997). One of thechannels through which a temporarily stronger exchange rate may influencethe growth rate is the impact on investment. For example, an overvaluationmight hurt investment, even though it lowers the price of imported capitalgoods, because it reduces the returns to investment in the tradables sectorand because the resulting current account deficit creates the need for tightermacroeconomic policies (Bleaney and Greenway, 2001).

When aid flows build up public infrastructure and thus augment the pro-ductivity of private factors, it is possible to realize significant medium-termwelfare gains from aid, even in the presence of some short-run Dutch disease(Adam, 2005). Bevan (2005) has suggested that the best practical approach isoften to ignore real exchange rate effects except when there is specific infor-mation on their likely magnitudes. This neutral assumption would be more


Real exchange rate appre-ciation can have a signifi-cant effect on exportgrowth in sub-SaharanAfrica.


appropriate in a scenario with a gradual scaling up rather than a very rapidincrease in aid inflows, and when the supply payoff (for example, from phys-ical infrastructure) is likely to be more rapid. As an example, IMF (2005a)found no evidence that aid flows to Ethiopia after 1991 (that is, following theoverthrow of the Derg regime) caused a real appreciation or harmed noncof-fee exports. Instead, foreign aid was found to have a positive impact both onEthiopia’s noncoffee exports and on their share in total exports.

Historical relationships might not be a reliable guide to the future, how-ever, and given that the resource flows required for SSA to achieve the MDGswill be significantly higher than past aid levels, there seems a strong likelihoodthat such a scaling up of aid will put upward pressure on wage and price lev-els and cause a real exchange rate appreciation. It would thus be prudent toimplement policies to counter such pressures. Building into the scaling-upscenario an assumption that the exchange rate will appreciate may also bewarranted if a high proportion of aid is spent on domestic goods (as discussedin Section 5.4). These judgments and assumptions of course must be tailoredto the specific country’s circumstances.


Scaling up aid to levelssufficient to achieve the

MDGs is likely to raisewage and price levels and

to lead to real exchangerate appreciation in sub-

Saharan Africa.


3Using Monetary Policy for Sterilization and

Inflation Management

3.1. Policy Options for Sterilization

If the monetary authorities are concerned with the liquidity impact ofincreases in aid-induced spending, they can sterilize the liquidity injectioneither domestically or through foreign exchange sales.

3.1.1. Domestic Sterilization Measures

There are several ways to domestically sterilize the liquidity impact of aidinflows, although one overall concern with this type of policy is the negativeimpact on private sector lending and investment. Three such measures arethe sale of government securities or sterilization bonds, increased reserverequirements, or a transfer of public sector deposits to the central bank fromcommercial banks.6 There are discussed in turn.

• Sales of government securities or central bank sterilization bonds:Monetary authorities often use this type of operation to sterilize the liq-uidity impact of capital inflows. One drawback, however, is that increas-ing the outstanding stock of domestic debt may lead to higher domesticinterest rates, especially in shallow financial markets. In addition toincreasing domestic debt service, sales of such instruments can crowdout loans to the private sector and, ultimately, investment. This policycan make domestic interest rates extremely volatile in the short term, asoccurred in Uganda.

• Increased reserve requirements: This can be an effective means ofmanipulating liquidity conditions, but there are some drawbacks. Some

16

Three measures for domesticsterilization of aid inflowsare sales of government securities or bonds, increasedreserve requirements, ortransfer of public sectordeposits to the central bank.

6Given that international capital mobility is highly limited in SSA countries, monetary policycontrols the nominal interest rates because there are no capital inflows that can offset the tight-ening of the monetary policy stance.


banks may already hold reserve assets in excess of statutory require-ments, which would reduce the impact of the policy change. The pres-ence of weak banks may make higher requirements difficult ordangerous to implement, especially if the remuneration of requiredreserves is significantly below market rates. Reserve requirements can-not be increased to manage short-term liquidity (which may be neededwhen aid flows are highly volatile) because frequent changes can under-mine the efficient management of bank portfolios. Finally, for countriestrying to liberalize their financial markets, changing reserve ratios oftensends the wrong signal and may discourage financial intermediation,weakening the central bank’s monetary control.

• Transfer of public sector deposits from commercial banks to centralbank accounts: This policy entails fewer fiscal or quasi-fiscal costs thanopen market operations, but there may be limited scope for such opera-tions in SSA countries where central government deposits in commercialbanks amount to about 2 percent of GDP, on average. Such a sterilizationpolicy was highly effective in Thailand and Malaysia in the 1990s.However, such a transfer should only be a one-time operation because ofconcerns similar to those related to changing reserve requirements: fre-quent movements of large amounts of public sector deposits may preventoptimal management of commercial bank portfolios.

3.1.2. Foreign Sterilization

Foreign sterilization involves selling central bank foreign reserves inorder to absorb the increased domestic liquidity. Some countries have used amixture of foreign and domestic sterilization. However, the increased supplyof foreign exchange creates pressure for a nominal appreciation of theexchange rate.7 Given concerns about competitiveness, authorities are oftenreluctant to accept such an appreciation and instead choose domestic steril-ization.

3.1.3. Domestic Versus Foreign Sterilization

There remains the question of when to use sales of domestic bonds andwhen to use foreign reserves to sterilize aid inflows. Atingi-Ego (2005) sug-gests a rough 50:50 split, but experience in Uganda seems to argue in favor ofsterilizing through foreign exchange sales, because exchange rate apprecia-tion has not hurt Uganda’s nontraditional exports. A similar response isemerging in Tanzania because of the need to balance the pressure on pricesfrom increased liquidity against the pressure both on interest rates fromdomestic sterilization and on exchange rates from increased foreign

Using Monetary Policy for Sterilization and Inflation Management 17

Foreign sterilizationinvolves selling central

bank foreign reserves toabsorb domestic liquidity.

7To mitigate appreciation pressures, the authorities can also relax controls on capital outflows,such as by easing surrender requirements on foreign exchange earnings or permitting local insti-tutions to invest abroad.


exchange sales. Finally, IMF (2005d) suggests that countries create steriliza-tion headaches for themselves by not allowing more aid absorption in the firstplace. It suggests that countries receiving higher aid flows should be morewilling to absorb (and, ultimately, spend) aid, selling the foreign exchangeover time and letting the exchange rate appreciate. However, the sale of for-eign exchange can reduce the share of domestic assets in private banks andmay crowd out private sector credit if lending in foreign currency is limited.

3.2. Managing Inflation

The scaling-up scenario needs to include the likely effects of increased aidflows on inflation. Some additional inflationary pressure is inevitable asdomestic demand increases. Even so, IMF (2005f) argues that although highinflation is clearly harmful to economic growth, the gains from continuedmoderate inflation on growth are ambiguous.8 The negative effect on growthof high inflation stems from the associated increase in inflation uncertainty,which clouds price signals and limits both the quantity and the quality ofinvestment. On the other hand, moderate inflation can enhance real wageflexibility. Furthermore, if nominal prices are inflexible, an excessively lowinflation target can render an economy more vulnerable to prolonged down-turns in case of adverse supply shocks. Therefore, an inflation target below 5percent may be not be appropriate.

Quantifying the association between inflation and growth requires care-ful attention to the nonlinearities in the relationship, which appear to be con-vex. That is, the lower the initial inflation level, the higher the negative effecton growth of a given increase in inflation; but at very low initial rates,increases in inflation may have either no effect or a positive effect on growth.Since Fischer (1993), several authors have tried to identify and locate such a“kink” in the relation between inflation and economic growth, which theyassociate with a maximum potential growth rate. Empirical studies using pan-els of countries locate this kink in the inflation-growth nexus at a level of infla-tion somewhere between 3 percent and 40 percent, with a majority suggestinga level in the 5–10 percent range (see Appendix Table A.2.3). In the Ethiopiascaling-up scenario prepared by the IMF (2005b, 2005c), the inflation targetin the base case projection was 3 percent, but was increased to 6 percent inthe “doubled aid” scenario.


Increased aid inflows will inevitably create inflationary pressure.

8The existence of a negative relationship between inflation and growth at higher rates of infla-tion is empirically well supported. By contrast, identifying the growth effects of moving from, say,20 percent inflation to 5 percent, has been challenging. According to Bruno and Easterly (1998),significant adverse growth effects can be found only for generally short-lived periods of highinflation, after which growth tends to return to its long-run path. However, several other studiesindicate this may understate the adverse growth effects of moderate inflation.


4Mobilizing Adequate Domestic Revenues

Any scaling-up scenario needs to take into account the possible effects ofaid on revenues. The associated policy package should stress the need tomaintain or strengthen revenues during the period of higher aid, both toguard against uncertain donor behavior and to prepare for an eventual taper-ing off of aid flows.

4.1. Preventing Aid Dependency

If policymakers treat external aid resources as substitutes for domesticrevenues, a substantial scaling up of aid could dampen a country’s domesticrevenue mobilization, specifically by weakening the tax collection effort. Insome cases, to the extent that a weaker tax effort reduces domestic distor-tions, it might spur economic activity. In other cases, where weaker revenuecollection reflects poor compliance or unnecessary tax exemptions, it wouldbe more likely to breed aid dependency. Furthermore, a weaker tax effortcan have an adverse effect on domestic institutions because citizens are lesslikely to hold the government accountable when they pay lower taxes (Bevan,2005). An argument can be made that reducing tax rates can be an optimalresponse to permanently higher aid flows, but this argument has less weightfor countries that are currently below their potential for raising tax revenuesand for which scaled-up aid inflows will be temporary. This latter group of countries must establish a strategy for coping with an inevitable drop inaid flows.

Aid flows historically have been both volatile and unpredictable, whichraises the concern that an increased reliance on external aid resources tofinance expenditures can constrain policymakers’ ability to undertakemedium-term planning. Bulir and Hamann (2005) estimate that, on average,aid flows are between 6 and 40 times more volatile than fiscal revenues andthat the relative volatility of aid is the highest for the most aid-dependentcountries. In addition to being volatile, aid flows are also unpredictable, espe-cially for the poorest countries. Bulir and Hamann (2005) calculate that, onaverage, aid delivery falls short of pledges by more than 40 percent.

19

Aid flows are most volatileto those countries most

dependent on them.


Aid volatility and unpredictability can be mitigated in various ways,including through debt relief, reserve buffers, and changes in donor behavior(Table 3). From the recipient country’s perspective, one clear advantage ofdebt relief over other forms of aid is that it permanently releases domesticresources. Debt relief thus mitigates the problem identified by Eifert and Gelb(2005): if aid pledges are highly volatile, recipients may slip into a low-levelequilibrium as they heavily discount commitments of aid in their medium-term budget plans, while donors, seeing lower funding gaps, may reduce theirpledges as a result. However, Eifert and Gelb (2005) also show that relativelysmall amounts of accumulated reserves (on the order of two to three monthsof imports) can be enough to manage much of the exogenous volatility of aid,especially that arising from administrative delays.9 More generally, aid shouldbe absorbed only if it is well spent and generates sufficient gains to compen-sate for any costs associated with a real exchange rate appreciation. Evenwhen the consequences of Dutch disease are limited, the need to insureagainst shocks, including future aid shortfalls, justifies some accumulation ofreserves when aid flows are considered high. Over time, however, there maybe a practical limit to the level of reserves, because donors may reduce aidflows if they observe that aid is not being absorbed by the economy.

4.2. Protecting Revenues

The empirical evidence on how aid flows affect domestic revenue collec-tion is mixed, with the magnitude, sign, and significance of the impact of aidvarying by study. With a few notable exceptions, however, the impact of aid isfound to be either negative or insignificant. Appendix Table A.2.4 summa-rizes these empirical findings. The composition of aid (loans or grants), thelevel of corruption, and the tax treatment of aid are important in assessing thelikely pressures on the revenue effort.


An increase in aid can befully offset by reduced rev-enues in countries with ahigh level of corruption orweak institutions.

Table 3. Mitigating Aid Volatility

Means Benefits

Debt relief There is a permanent release of domestic resources.The level of aid is predictable, which facilitates medium-term budget planning.

Reserve buffers Relatively small reserves can smooth aid flows, includingadministrative delays, and insure against shocks.

Changes in donor behavior Long-term aid pledges and more regular disbursementof aid commitments can improve recipient countries’ability to effectively spend aid, including by reducingtheir need to accumulate reserves.

9In addition, donors are placing increased emphasis on evaluating their own performance inaid management. For the case of Mozambique, this is manifest in an agreed performance assess-ment framework for the donor community (World Bank, 2005).


4.2.1. Loans Versus Grants

Gupta and others (2004) suggest that the need to repay loans leads poli-cymakers to increase their domestic revenues, or at least to maintain theirexisting collection levels. Grants are free resources that can substitute fordomestic revenues, and hence are more likely to dampen domestic efforts tocollect more revenue. Their baseline results indicate that 28 percent of everyadditional dollar of aid in the form of grants is offset by reduced domestic rev-enues.10 Their results suggest that for countries with a high level of corrup-tion, any increase in aid is fully offset by a reduced revenue effort.

4.2.2. Tax Treatment of Foreign Aid

When aid consists of foreign-financed projects, their tax treatment pre-sents an additional complication. If the projects are tax-exempt, then the fis-cal component of the scenario must be adjusted accordingly. However, taxexemptions can be difficult to manage when a country’s administrative capac-ity is weak. Some SSA countries have had success in implementing exemptionsthrough vouchers rather than administrative provisions (de Mariz Rozeira,2005).11 However, a case can be made, on grounds of both simplicity and fis-cal revenue protection, for having donors pay all applicable taxes and dutieson foreign aid projects. These decisions have to be examined on a country-by-country basis.

Experience points to a tax ratio of at least 15 percent of GDP as a rea-sonable target for most low-income countries.12 At the same time, donorsmust bear in mind the composition of tax revenues in scaling-up scenariosbecause on average countries in SSA derive about one-third of their total rev-enue from trade taxes (Agbeyegbe, Stotsky, and WoldeMariam, 2004). If aidrecipients liberalize trade to facilitate aid absorption (as discussed in Section2.2), revenue from trade taxes would decline unless there were scope for elim-inating exemptions and reducing tariffs to revenue-maximizing levels. Theywould also have to strengthen their indirect tax system, including the value-added tax (VAT), by reducing exemptions and broadening the base. Thesecountries would have to build transitional revenue losses into their scaling-upscenario and would need donor and other assistance to improve domestic taxsystems and recoup losses from trade taxes.

Mobilizing Adequate Domestic Revenues 21

Scaling-up scenarios mustincorporate the tax impli-

cations of aid, includingtax treatment of aid-financed projects andeffects on trade taxes.

10Their sample includes 107 developing countries from 1970 to 2000. Otim (1996), in his studyof three low-income South Asian countries, finds that both grants and loans increase revenue.

11Under the voucher system, normal tax and customs procedures apply to projects that receivetax incentives. Instead of payment, the beneficiary presents a voucher or treasury check issuedby the ministry of finance to the relevant customs or tax office. The associated paper trail makesmonitoring and costing much easier than with a tax exemption.

12Adam and Bevan (2004) speak of a consensus that the tax ratio for post-stabilization coun-tries should be on the order of 15–20 percent, and IMF (2005f) suggests a ratio of at least 15 per-cent as a reasonable target for most low-income countries.


4.2.3. Revenue Assumptions for Scaling-Up Scenarios

The most appropriate assumption for scaling-up scenarios is that accom-panying policies ensure that the revenue effort is strengthened, possiblythrough the use of revenue benchmarks, particularly in those countries wherethe revenue-to-GDP ratio is well below the threshold of 15 percent. A strength-ened revenue effort would enable aid-dependent countries to wean them-selves gradually from aid. It is also consistent with the recommendation of theUN Millennium Project (2005) that countries mobilize additional domesticresources of 4 percent of GDP through, for example, more vigorous tax col-lection efforts. In many countries, the tax effort is below potential, in partbecause of a narrow tax base. Some countries, including Tanzania andUganda, have emphasized broadening the tax base with the objective ofreducing aid dependency over time. McGillivray and Morrissey (2001) notethat it is useful to distinguish between the impact of aid on total tax revenueand the impact of aid on revenue from international trade taxes. AppendixTable A.2.5 summarizes recent trends in total tax revenue and import andexport duties in different regions.

Scaling-up scenarios should also factor in support to help revenue insti-tutions strengthen their capacity to generate additional revenue on a perma-nent basis. Addressing weaknesses in revenue administration is high on thereform agendas of Tanzania and Uganda, for example.


Policies to strengthen therevenue effort enable aid-dependent countriesto wean themselves off aid.


5Projecting the Impact of Increased Aid on

Economic Growth

5.1. The Relationship between Aid and Growth

The debate about the effectiveness of aid in stimulating growth goes backmany years, yet there remains considerable uncertainty about the aid-growthrelationship. Some researchers suggest that there is either no effect or a neg-ative one; others suggest a positive effect, but with diminishing returns. Stillothers argue that aid works to promote growth in some circumstances (whena country has good policies), but not in others.

Early skepticism about the role of aid in promoting economic growthfocused on the potential disincentive effects of aid on investment and pri-vate sector development (Bauer, 1972). Some of the ensuing empiricalresearch did in fact find little or no relation between aid and growth(Mosley, 1980; Singh, 1985). Similarly, Easterly (2001) questions the chan-nels through which the “financing gap” model purportedly promotesgrowth, specifically whether aid flows lead to higher investment levels andhence to growth.13 There is evidence to indicate that, although aid can ease the liquidity and foreign exchange constraints to investment, it mayactually worsen the incentive to invest, and hence be more likely to financeconsumption.14

A related but distinct strand of the literature examines the impact of aidflows on the tradable goods sector. Rajan and Subramanian (2005a) find thataid flows do have adverse effects on growth, wages, and unemployment inlabor-intensive and export sectors. However, their model does not allow forinferences regarding the overall growth rate.

23

There remains great uncertainty about the

relationship between aid and economic growth.

13The “financing gap” or alternatively the “two-gap” model by Chenery and Strout (1966)identifies the gaps between savings and investment requirements and between foreignexchange earnings and import requirements as the major constraints to growth in developingcountries.

14Empirical support for this idea can be found in Boone (1996).


The effectiveness of aid can be affected by the policy environment in therecipient country (Burnside and Dollar, 2000) and by the quality of gover-nance (Radelet, 2004). Other studies have examined whether the impact ofaid is conditional on other factors.15

5.1.1. The Impact of Different Types of Aid

Different types of aid are likely to have different relationships with growth, and not all aid is targeted at stimulating growth (Clemens,Radelet, and Bhavnani, 2004). A distinction can be drawn between threetypes of aid:

• Aid provided in the context of disasters, emergencies, and humanitar-ian relief (for example, food aid), which may actually have a negativecorrelation with observed growth rates because it is often providedwhen a country is hit by a negative shock.

• Aid that might affect growth but, if so, only indirectly and over a long time: Included in this group is aid to halt environmental degra-dation or to support democratic or judicial reform, and some aid tosupport health and education, which all may have an effect on laborproductivity only over many years (Clemens, Radelet, and Bhavnani,2004).

• Aid that might reasonably be expected to affect growth within a fairlyshort time (for example, four years), including aid to build infrastruc-ture such as roads, irrigation, ports, and electricity.

5.1.2. The Growth Effects of Scaling Up

To project the impact of scaling up on real growth rates, therefore, onemight start with how the aid will be spent and the policy environment intowhich it will be disbursed. It is important to categorize the projected aidinflows and the associated higher budget expenditures into 1) those that canreasonably be expected to enhance growth in the short to medium term, 2)those directed at longer-run growth, and 3) those that support activitiesunlikely to be related to growth.

Cross-country econometric studies, which often use five- or ten-year aver-ages of growth rates, may not provide sufficient guidance on the short-runimpact of higher public spending on output in individual countries.Moreover, most empirical work on the impact of spending on growth providesestimates of the average impact of small changes in spending as a share ofGDP, which may not directly translate into the impact of a substantial scaling


Aid inflows should be categorized according towhether 1) they can reasonably be expected to enhance growth in theshort to medium term, 2) they are directed atlonger-run growth, and 3) they support activitiesunlikely to be related togrowth.

15These include the effects of large export price shocks (Collier and Dehn, 2001), climaticshocks and trends in or the volatility of terms of trade (Guillaumont and Chauvet, 2001; Chauvetand Guillaumont, 2002), and the presence of totalitarian regimes (Islam, 2003).


up of aid, in which diminishing returns and other supply bottlenecks must fig-ure more prominently.

5.2. Accounting for Diminishing Returns and Limits toAbsorptive Capacity

Projecting the total response of output to scaled-up public expenditurerequires accounting for diminishing returns to spending, the pace of conver-gence of output to its new steady state, and supply constraints and bottle-necks. An underlying notion is that the link between spending and growthreflects intermediate outputs (such as education or health capital or publicinfrastructure), the production of which can face constraints. Assessing thesepotential constraints in individual sectors is key to preparing a realistic assess-ment of the impact of scaling up.

Information on the extent of diminishing returns and the rate of conver-gence is subject to considerable uncertainty. Guidance is available on the ini-tial response coefficient, but there are few estimated production functions forgovernment spending or aggregate output in low-income countries that couldinform the selection of the diminishing returns and convergence parameters.

5.2.1. Convergence Parameters

Existing estimates of convergence parameters are averages from cross-country studies and would have to be adjusted on a country-by-country basisto take account of local conditions and policy assumptions. Other thingsbeing equal, low-income countries have more scope to catch up with richercountries and should therefore be able to maintain a growth rate higher thanthe steady state for a long time. However, studies that control for other causesof growth (for example, strength of institutions) suggest that many low-income countries may be close to their steady state already. Therefore, con-vergence in response to a policy change may be more rapid, and the long-runimpact of aid on growth may be correspondingly muted. Hence, measuresthat raise the steady-state level of income, such as strengthening governanceand building up institutions, are important for ensuring that increased spend-ing leads to sustained growth.

5.2.2. Diminishing Returns to Aid

The aid saturation point is the point at which the positive impact of aidfalls to zero. This varies widely in different studies. The diminishing returns toaid are often captured in the empirical literature through the inclusion in anaid impact regression of a quadratic aid share variable along with the stan-dard linear term. Clemens and Radelet (2003) summarize eight such studiesrelating the share of aid to per capita growth. All find a negative coefficienton the quadratic aid term, which, when combined with a positive coefficienton the level of aid, implies that the marginal return on aid is initially positive

Projecting the Impact of Increased Aid on Economic Growth 25

The link between spending and growth

reflects intermediate outputs that are subject

to diminishing returns,supply constraints, or

other bottlenecks.

The aid saturation point—when the positive

impact of aid falls tozero—is higher in

countries with a good policy environment.


but then declines.16 The growth effects of most social sector expenditures,which have a long-term impact on growth, are more difficult to estimate. Inaddition, all existing saturation point estimates are derived from historicaldata, and saturation points in the future may be at significantly higher levels,particularly if aid absorption is accompanied by improvements in the policyenvironment and governance.

There is evidence that countries’ aid saturation points and absorptioncapacities are higher in good policy environments.17

5.2.3. Postconflict Situations

The sequencing and composition of aid require special attention in post-conflict countries. These countries face more severe constraints than the typicalrecipient of increased aid: basic institutions have to be rebuilt before attentioncan turn to achieving the MDGs. If rebuilding is successful, growth in thesecountries can rebound rapidly with restoration of law and order and the returnof dislocated people. Aid is effective in these circumstances, but to be optimal,should change in composition over time (Chauvet and Collier, 2004).Furthermore, donors tend not to time aid properly—aid tapers off after threeto five years, just when a country’s opportunity to build capacity and achieve sus-tained poverty reduction is best (Collier and Hoeffler, 2004). The amount of aidrecommended for postconflict countries is quite similar to the level of a scaling-up program. This suggests that, after the first stage of rebuilding is complete,the guidelines in this paper are directly relevant for postconflict countries.

5.3. Safeguarding Private Savings and Investment

Public investment can crowd in private investment in SSA countries.Crowding in likely reflects the complementarity of private investment with


The basic challenges of a scaling up of aid are relevant to postconflict countries, after the initial rebuilding is complete.

16While considerable caution needs to be exercised in using these results because they are ten-tative, Hansen and Tarp (2001) show aid share saturation points between 14 and 27 percent ofnational income. Other studies find that saturation occurs at about 40 percent, well above thelevel of current aid inflows for most countries. For their short-impact aid component, Clemens,Radelet, and Bhavnani (2004) find the marginal impact of aid reaches zero at about 8–9 percentof GDP. But since short-impact aid is about half of total aid, the corresponding point for total aidoccurs at about 16–18 percent of GDP, a range currently exceeded in only a small number ofcountries. Aid saturation points will be lower for postconflict countries, especially in the initialstages of reconstruction, but these aid flows will themselves help raise absorptive capacity overthe medium term (as discussed in Section 5.2.3).

17Several of the studies surveyed in Clemens and Radelet (2003) allow the impact of aid ongrowth to depend on an indicator of the quality of the country’s institutions and policy stance.Collier and Dollar (2002) model a link between aid impact and the World Bank’s Country Policyand Institutional Assessment index, which ranges from 1 (lower quality) to 6 (higher quality).For a country with a score of 2, the saturation point is about 19 percent of GDP. With a score of4.5, the saturation point occurs at 43 percent of GDP. This result is consistent with other evidencethat the effectiveness of public expenditure depends on the quality of institutions (Baldacci andothers, 2004). More recent evidence in Rajan and Subramanian (2005b) is more cautious aboutthe existence of such a link.


some components of public investment, especially infrastructure (Odedokun,1997). Evidence on crowding in has been found across a variety of datasetsand methodologies.18 Appendix Table A.2.6 summarizes some of the relevantstudies.

Strengthening the investment climate and the financial sector are impor-tant elements of any scaling-up scenario. In recent years, many low-incomecountries have prepared Diagnostic Trade Integration Studies (DTISs) as ameans of integrating trade and investment issues into their developmentstrategies.19 For example, the 2004 DTIS for Ethiopia recommended improve-ments in trade policies, the legal and regulatory environment, institutions,and trade-facilitation services, in order to encourage greater integration intothe world economy and increased foreign direct investment.

More generally, measures should be included in every scaling-up scenarioto strengthen financial sector institutions in order to raise private sector sav-ings and investment over the medium term and to facilitate the country’seventual graduation from relying on official sources of finance to relying onprivate sources. These measures should include instituting international stan-dards for bank supervision, a competitive environment, and well-functioningfinancial markets, as well as taking steps to enhance the microfinance sectorand to improve access to credit.

5.4. Raising Spending as a Share of GDP

Econometric evidence on the initial impact of public spending on growthis sensitive to the data and the methodology used. The link is clearest whenpublic spending can be related to the stocks of the factors of production, suchas physical capital, which are augmented by public investment. Many studiesconfirm a productive role for various types of infrastructure in low-incomecountries. Appendix Table A.2.7 summarizes studies that use either the pro-duction function or growth regressions and draw on data from Africa.20

Output responds positively to infrastructure, and there are also strong com-plementarities between different components of capital spending (Canningand Bennathan, 2000). The implication is that rates of return decline veryquickly for an increase in any one component of capital. For low-incomecountries, electricity projects may yield the highest returns (Canning and


Raising private sector savings and investment

over the medium term will facilitate a country’s

eventual graduation from relying on official

sources of finance.

18Greene and Villanueva (1991) note this effect for a small sample of developing countries inSSA, Asia, and Latin America, using a pooled approach. Hadjimichael and Ghura (1995) find itin a panel regression for a large SSA sample, while Ghura and Goodwin (2000), also using panelmethods, find that the effect is present only for SSA, while crowding out was observed for Asiaand Latin America. Belloc and Vertova (2004) investigate the effect at the country level and con-firm crowding in for four of the five SSA countries they examine (the exception being Malawi).

19The Integrated Framework (IF) which sponsors the DTIS, was established in 1997 by six mul-tilateral institutions (IMF, ITC, UNCTAD, UNDP, World Bank, and WTO) in order to facilitatethe coordination of trade-related technical assistance to low-income countries.

20Of the individual developing country studies reported in Briceño-Garmendia, Estache, andShafik (2004), all find that infrastructure has a positive effect on output, although none coversAfrica specifically.


Bennathan, 2000). There may also be large gains in growth from closinginfrastructure gaps between average countries and regional leaders (Calderonand Serven, 2004).

Growth regressions find that the average impact on per capita growth ofan increase of 1 percent of GDP in social sector or public investment spend-ing is in the range of 0.5 to 1.0 percent over a five-year period. These studiesare summarized in Appendix Tables A.2.7 and A.2.8.21

5.5. Confirming the Positive Impact of Aid

Case studies show that aid has a positive impact on growth, although it isdifficult to precisely correlate the causes and effects given the variationsamong countries’ circumstances and their use of aid inflows. The multidonorPro-Poor Growth study (Agence Française de Développement and others,2005) finds that aid inflows fostered higher growth in the 1990s for theAfrican case study countries. The underlying studies focus both on the impactof aid on growth and the ability of the poor to participate in growth. Aidimpact on growth was strongest in Uganda, operating through reconstruc-tion, improved economic management, social sector programs, and improve-ments in public administration. Aid also played an important role in relaxingconstraints on growth due to debt burdens. Similarly, the Ghana case studyfinds that aid played the twin roles of supporting macroeconomic stabilizationand boosting social sector programs that might otherwise have been cut dueto scarce resources. The other case studies (Zambia, Burkina Faso, andSenegal) explore the impact of aid in less detail, but all note the importanceof aid for financing health and education programs and thus supporting thehuman capital component of the growth process.

The strength of public expenditure management (PEM) systems is alsoan important determinant of the growth impact of aid inflows. Studies of thefiscal impact of aid in Uganda, Zambia, and Malawi by the OverseasDevelopment Institute (Fagernäs and Roberts, 2004) find the strongest dis-cernible growth impact of aid in Uganda. For the other countries, the mainbenefit was the ability to protect some critical social sector and public invest-ment programs at a time of overall fiscal stringency and policy slippage. Oneclear difference among the three countries was in the performance of the aidallocation mechanisms. Uganda comes much closer to a single integrated


Public spending has agreater impact on growthwhen it is directed towardthe factors of production,such as physical capital,which are augmented bypublic investment.

Aid can protect social sector programs and public investment in timesof scarce resources or policy slippage.

21The methodological debate associated with these studies has confirmed the need to controlfor reverse causality from output growth to public investment. Of particular relevance to SSA isthe study by Gupta, Clements, Baldacci, and Mulas-Granado (2004) which covers 39 IMF pro-gram countries, of which 24 are in SSA. It finds a 0.7 percent effect on the growth rate of a 1 per-cent of GDP increase in capital outlays over a five-year period. In a larger sample of 120countries, Baldacci and others (2004) simulate the increment to long-run growth from increasesof 1 percent of GDP in education and health spending. For education, per capita GDP growthis, on average, 0.9 percent higher, and, for health outlays, it is 0.4 percent higher. These estimatesare derived from the experiences of countries in which the elasticity of imports is less than one,and, hence, there is less than full leakage of higher aid flows to imports.


budget framework for all funding sources, allowing aid to be directed to itsmost productive use and accentuating the growth impact. In Malawi andZambia, however, there persists a traditional assignment of aid inflows to spe-cific development programs, even when these did not offer the highestgrowth impact.

Countries’ experiences with resource windfalls provide additional lessonsfor scaling-up countries, specifically by pointing to the need for effectiveexpenditure management and for targeting of aid inflows. The analogy of aidinflows to natural resource inflows is imperfect, but both inflows generate sim-ilar policy challenges, including the risks of Dutch disease, misappropriation,and institutional deterioration resulting from diminished revenue mobiliza-tion incentives (Hausmann and Rigobon, 2003).22 Resource windfalls indeveloping countries can lead to a dissipation of resource wealth and evennegative growth. This reinforces the importance of aid being delivered withina policy framework supportive of the efficient use of the additional resources.


Aid is most effective when it is delivered

within a policy frameworkthat supports efficient use of the additional

resources.

22The analogy of aid inflows and natural resource booms is imperfect because aid is normallyprovided by donors after agreement on appropriate policy conditions. Moreover, if it becomesclear that aid inflows are not being fully absorbed, they are likely to diminish.


6Meeting Other Fiscal Challenges

Fiscal policy can become more complicated in a high-aid environment(Heller, 2005), and this section considers a number of fiscal issues that mayneed to be considered in a scaling-up scenario.

6.1. Developing an Exit Strategy

When a government scales up its expenditure program, it generally hiresmore workers, delivers additional services to the public, and has new infra-structure to maintain. The government then faces the challenge of what to doif, or most likely when, donors do not sustain aid at the higher level. If itbecomes difficult to reduce expenditures in response to a gap or reduction inaid financing, the pressure may increase significantly to domestically financethe deficit.

This underlines the importance of preparing an “exit strategy,” that is, themacroeconomic path the country will follow after scaled-up aid flows fall backto more normal levels or, perhaps to lower-than-normal levels, if aid flows havebeen front-loaded. If a government chooses to smooth the impact of aidvolatility, for example, it might plan to accumulate a certain amount ofreserves (or use some of its existing reserves), and this will affect monetarymanagement strategies (as described in Section 3).

6.1.1. The Scaling Down of Aid

Planning for an end to increased aid inflows is a critical element of anyscaling-up scenario. The scenario assumes a temporary spike in aid inflows,for example, over a 10-year period, followed by a corresponding “scaling-down” period, for example, the subsequent 10 years, during which aid flowsreturn to historical levels. It is important to illustrate the potential impact ofthe scaling-down period on the macroeconomic projections, even thoughthese outer-year projections will be highly uncertain.

The temporary increase in aid allows the government to increase currentand capital spending as a share of GDP, but policies during the scaling-up

30

The exit strategy is themacroeconomic path thecountry will follow afterscaled-up aid flows fall backto more normal levels.


period must be consistent with a smooth transition during the scaling-downperiod to accommodate a lower aid environment. As noted, if the governmentfinds it difficult to reduce expenditures from previously aid-financed levelsduring the scaling-down period, there may be significant pressure for higherdomestic financing of the deficit.

6.1.2. Private Economic Activity

Private sources of finance must eventually substitute for official aid flowsin stimulating economic activity. The temporary nature of the exceptional aidflows underlines the need to maintain or enhance the domestic revenueeffort (see Section 4), and to encourage, over time, higher private sector sav-ings and investment, including policies to stimulate foreign direct investment.A strong private sector response (and hence a real expansion of the tax base)will lessen the fiscal adjustment required when external aid is reduced. It willalso reduce any negative effects of scaling down on growth.

6.1.3. Real Spending Levels

In preparing the exit strategy, it is likely that public current and capitalspending will be assumed to fall as a share of GDP. However, it is importantthat this scaling down of aid is sufficiently gradual to allow current spendingto be maintained in real terms, if possible. Building up a higher level ofreserves during the scaling-up period may also allow for a more gradualreduction of public expenditures during the scaling-down years.

6.2. Properly Estimating Current Spending

Countries should incorporate into their MTEFs adequate current expen-ditures to support the projected increases in investment over the mediumterm. A scaled-up public investment program is likely to encounter highmarginal recurrent costs even if average current costs are low.23 A scaling-upscenario should include generous estimates for current expenditures on newfacilities, particularly when current expenditure is already inadequate toproperly operate and maintain existing facilities and is not part of the MTEF.The factors contributing to the tendency to neglect current expenditure arewell known and include donors’ past preference for capital projects, the per-ception that political advantage attaches to new projects, and the lack ofimmediate negative consequences from postponed maintenance.

Meeting Other Fiscal Challenges 31

Policies during thescaling-up period must beconsistent with a smooth

transition during thescaling-down period.

23 Hood, Husband, and Yu (2002) calculate average proportions of annual recurrent costs toinvestment expenditure for World Bank and Asian Development Bank projects across a range ofsectors. The proportions were up to 7 percent in health, 5 percent in transport and energy, and33 percent in education, but there was also enormous variation even for similar types of projects.Certain types of variation were straightforward to explain; for instance, upgrading existing facil-ities imposes lower current costs than building new facilities because the recurrent obligations ofthe former were in place before the project began.


Governments have found it challenging to mobilize domestic resourcesfor operations and maintenance (for example, through user charges), espe-cially when the service may be considered essential. On the other hand, theexistence of underused facilities offers the prospect of fairly quick returns toadditional current spending, as long as these facilities can be rehabilitated forless than the cost of constructing new facilities. Over the longer term, gov-ernments must strengthen their budgeting systems to incorporate reliableestimates of current spending requirements.

6.3. Targeting the Poor

Sub-Saharan African countries have raised social sector spending in recentyears, but this spending is not always well targeted toward poor households. Arecent review of the experience of 56 countries from 1960 to 2000 finds that thetargeting of social sector spending toward the poor in SSA was very ineffective,although there was some improvement in the 1990s (Davoodi, Tiongson, andAsawanuchit, 2003). This is particularly noticeable for spending on education.It is fairly well known that spending on secondary and tertiary education tendsto benefit richer households, but SSA does not compare well with other regionseven in targeting spending on public primary education, of which only 18 per-cent goes to the lowest quintile of the population. Similarly, there is a wide gapin the benefits to the poorest and the richest households for public healthspending in SSA, especially for hospital spending, of which the poorest quintilereceives 12 percent of the benefits and the richest receives 31 percent.

6.4. Containing Unproductive Spending

There is a positive association between health and education outcomes, gov-ernance, and targeting effectiveness, suggesting that equity and efficiency com-plement each other and improve the overall impact of public spending(Davoodi, Tiongson, and Asawanuchit, 2003). This underscores the importanceof putting in place policies to appropriately target spending at poverty outcomesand to improve effectiveness and efficiency—that is, to reduce unproductivespending. “Unproductive spending” is defined as the difference between actualspending on a program and the reduced spending that would yield the samesocial benefit with maximum cost-effectiveness (see Chu and others, 1995).

6.5. Balancing Poverty Reduction and Growth

There may be a trade-off between poverty reduction or sector-specificgrowth and overall growth rates. Aid may have the highest return and pro-mote the highest growth rate when it is used to enhance the supply responseof the nontraded sector (Adam and Bevan, 2003). However, such a policy maynot be “pro poor” if the incomes of the poor are not closely linked to the tar-geted sector.


Social sector spending insub-Saharan Africa hasnot been well targetedtoward the poor.

Unproductive spending isthe difference betweenactual spending on aprogram and the reducedspending that would yieldthe same social benefitwith maximum cost-effectiveness.


Higher aid also can lead to wage pressures, which can have a negativeimpact on the poor who do not benefit from the wage increases. Specifically,social sector wages may rise when they are financed by aid, putting upward pres-sure on formal sector wages elsewhere in the economy, including manufactur-ing. To the extent that most poor people are dependent on agriculture and theinformal sector, such wage increases do not directly raise their income. In addi-tion, even though some poor people may benefit from increased formal sectorwages, they often have strong links to the export sector (Rajan andSubramanian, 2005a). Thus, social sector spending that is intended to reducepoverty may not be the optimal growth strategy or a pro-poor strategy, especiallyif increased social spending is not well targeted to the poor.

6.6. Minimizing Bottlenecks and Improving Coordination

Sectoral bottlenecks are difficult to anticipate and are best identifiedthrough frequent monitoring of aid impact indicators. Ideally, aid allocationis guided by an in-depth diagnostic assessment of sectoral capacity, with thegoal of avoiding excessive adjustment pressure in any one sector. A sectoralinvestment plan prepared by the World Bank and other donor partners canprovide guidance on, for example, projected employment needs in educationor health to meet the MDGs for these sectors. For instance, the World Bank’sMAMS model suggests that Ethiopia can achieve MDGs for universal primaryeducation, but that doing so would require 52,000 more teachers (a 66 per-cent increase) even at the current student-teacher ratio of about 75 to 1. Toraise the quality of education by reducing the ratio to 40 to 1 would require160,000 more teachers—a doubling of the current total. Such a large relativeexpansion in the number of teachers could trigger a substantial increase inwages or a decline in average skills.

6.6.1. Bringing NGO Activities into the Budget

Special challenges can arise when donors channel aid through nongovern-mental organizations (NGOs). Relying on NGOs can be useful if capacity withinthe public sector to absorb scaled-up aid is impaired. However, shifting alloca-tions to NGOs does not in itself avert macroeconomic absorption constraints,which relate to aggregate supply responses in the traded and nontraded goodssectors. Furthermore, it is important to ensure that NGO spending plans arecoordinated with those of the government budget to prevent duplication.

Channeling aid through NGOs also carries the risk of detracting fromefforts to build capacity for social sector delivery within the public sector. 24 If

Meeting Other Fiscal Challenges 33

Higher aid flows can leadto wage pressures, which

can have a negativeimpact on the poor.

Shifting aid allocations toNGOs does not in itself

avert macroeconomicabsorption constraints.

24For example, the United States established the President’s Emergency Program for AIDSRelief (PEPFAR), which will allocate $15 billion over five years (beginning in 2004) with spend-ing increasing as implementation capacity improves. The World Health Organization (WHO) issupporting the global “3 by 5” initiative which aims to supply antiretroviral drugs to 3 millionadditional HIV-positive persons by the end of 2005. A significant proportion of the funds fromboth programs will flow through NGOs.


poor countries are unable to absorb aid anywhere in their domesticeconomies in the short term, donors can divert part of the aid to financingglobal public goods (for example, malaria research) outside the recipientcountry. This would prevent the macroeconomy of the country receiving aidfrom experiencing adverse effects.

6.6.2. Government at Lower Levels

Strengthening the capacity of subnational governmental institutions isessential in countries where a substantial proportion of social sector spendingis devolved to lower levels of government. For example, Uganda has devolvedthe provision of most social services to the local governments. For this pur-pose, about 40 percent of the central government budget is transferred to thehighest local government tiers. However, the central government’s ability tomonitor the use of these resources remains weak. Therefore, scaling-up sce-narios need to assess the ability of the subnational governments to executeand report social sector spending and the capacity-building needs at differentlevels that can be supported by donors.

6.6.3. Existing Private Sector Capacity

When the public sector’s capacity to expand service delivery is limited, itcan be useful to build on existing private sector resources. In many cases, theprivate sector can expand more rapidly than the public sector to meet ascaled-up demand for services, further ameliorating sectoral capacity con-straints. Transfer schemes such as vouchers may be used to ensure that thepoor have access to basic services as private delivery expands.

However, the poor are often heavy users of private services in health andeducation because of failures in public service delivery to poor areas. Forinstance, the World Bank (2004) shows that private expenditures on health inSSA are as large as public expenditures and that the richest households in theregion had higher proportions of attended births and respiratory diseasetreatment in public facilities than poorer households. The public-private mixin health differs by type of service, with public facilities generally more impor-tant for preventive care like vaccinations and private facilities more importantfor treatment. The private sector also plays a significant role in education inSSA, but less so in utilities and infrastructure.


The poor are often heavyusers of private sectorservices in health andeducation.


7Strengthening Governance

There is a growing consensus that good governance is essential if higheraid flows are to be effective in promoting growth and reducing poverty.25

Major donors have made this point, most recently at the July 2005 summit ofthe Group of Eight countries in Gleneagles, Scotland (Group of Eight, 2005).Strengthening governance therefore will likely increase the probability thatdonors will actually disburse the higher aid flows promised in the new scaling-up programs.26

7.1. Reducing Corruption

Rent-seeking behavior, which is found in all countries, is more pervasivein some than in others. Pervasive corruption tends to be associated withpoorly enforced property rights, a weak rule of law, and weak incentives forproductive investment, all of which are damaging to economic growth ontheir own.

A considerable amount of research in recent years has focused on the neg-ative correlation between the level of corruption in a country and its social andeconomic performance. Corruption has received particular attention becausethe availability of corruption measures helps to “quantify” its extent and allowsfor international comparisons. Many studies rely on corruption indices devel-oped by Business International, International Country Risk Guide, andTransparency International. But two sets of World Bank indicators have gainedprominence in recent years: the Country Policy and Institutional Assessment

35

Reducing corruption can enhance growth andimprove aid absorption.

25See, for example, Commission for Africa (2005).26The link between aid and governance is a central feature of the U.S. Millennium Challenge

Account (MCA), which channels aid directly to specific country priorities. Country eligibility foraid is determined by a set of indicators, including one for corruption. For 2005, seven SSA coun-tries are eligible: Benin, Ghana, Lesotho, Madagascar, Mali, Mozambique, and Senegal. Sevenadditional SSA countries have “threshold” status, meaning that they are close to meeting the eli-gibility criteria and can access some aid flows to make additional progress. These countries areBurkina Faso, Kenya, Malawi, São Tomé and Príncipe, Tanzania, Uganda, and Zambia. Seehttp://www.mca.gov/.


http://www.mca.gov/.

(CIPA) scores (which became public information beginning in 2005), and theWorld Bank Institute’s indicators on six components of governance (whichcover nearly all countries and have been available biannually since 1996).

7.1.1. The Impact of Corruption on Economic Performance

Empirical research highlights the negative impact of corruption ongrowth, public finances, poverty, income inequality, and the provision ofsocial services. Appendix Table A.2.9 summarizes some of the results.27

Corruption is shown to lower growth by reducing private investment, attract-ing talented individuals into unproductive activities, and encouraging poormanagement of natural resources. Experience in the former Soviet Unionand Eastern Europe suggests that structural reforms designed to rationalizethe role of the state, increase reliance on market-based pricing, and create asound regulatory environment contribute to growth directly and indirectly bylowering the incidence of corruption (Abed and Davoodi, 2002).

Corruption also distorts the composition of public expenditures in favorof sectors in which bribes are easier to collect. Corruption typically shiftsspending away from routine maintenance and repair, education, and healthto excessive and inefficient public investments and higher military spend-ing.28 Human capital and investment are similarly impeded by poor gover-nance, which limits the growth impact of social sector spending. Corruptionin the form of abuse of public funds not only results in weak social indicators,but also weakens revenues because it contributes to tax evasion, improper taxexemptions, and weak tax administration (Ghura, 2002).

7.1.2. Successful Anticorruption Strategies

In a country with weak governance, effective anticorruption measuresshould strengthen the growth impact of a scaling up of aid. Successful anti-corruption strategies are typically predicated on the presence of a real andeffective deterrent to curb individuals’ instincts to abuse their public office forpersonal gain. Public officials need to believe that if they abuse their offices,they run a substantial risk of being caught, convicted, and punished.

Moreover, prosecuting serious high-profile corrupt actors is “an essentialelement of an anticorruption strategy so that a cynical citizenry believes thatan anticorruption drive is more than just words” (Klitgaard, Maclean-Abaroa,and Parris, 1999). Independent anticorruption bodies may have a role to play


Successful anticorruptionstrategies depend on thepresence of a credibledeterrent to the abuse ofpublic offices.

27A number of studies explicitly examine the negative impact of corruption on economicgrowth. Mauro (1995) finds that increasing corruption by one unit (on a scale of 1 to 10) lowersreal per capita GDP growth by 0.3 to 1.8 percentage points; Leite and Weidman (2002) and Abedand Davoodi (2002) report a somewhat narrower range centered on about 1 percent.

28Gupta, Davoodi, and Tiongson (2002) find that higher corruption has adverse consequencesfor social indicators such as child mortality rates and student dropout rates. Baldacci and others(2004) identify a key role for governance in influencing the effectiveness of education andhealth interventions. In particular, they find that health spending has no impact on social indi-cators in countries suffering from poor governance.


in this context. Their work should be supplemented by an ongoing publicinformation program to educate the public about the negative effects of cor-ruption and by a system that allows the public to report acts of corruptionwithout fear of retaliation.

It is also critically important to liberalize and reform institutions and prac-tices to reduce the opportunities for rent-seeking and corruption and tostrengthen public audit functions. These reforms also strengthen domesticrevenue mobilization by signaling the ability and commitment of the govern-ment to account for how revenue is spent.

7.1.3. Breaking the Cycle of Poor Governance

Beyond the negative impact on growth, there is a self-reinforcing cycle ofpoor governance. Citizens demand less and less from their government astheir demands go unmet and their expectations diminish. Breaking this cycleis an important element of the scaling-up agenda, although its contributionto improved governance is difficult to quantify.

7.2. Improving Public Expenditure Management Systems

Well-functioning PEM systems are essential if higher aid flows are to beabsorbed effectively. They provide assurance to donors that their resourcesare being used for the intended purposes, while reducing the transactioncosts of meeting donor-specified reporting requirements. They also improvegovernance by making public expenditures more efficient and transparent(including to citizens). Finally, they help governments implement the scaling-up scenarios by tilting expenditures toward priority areas.

Most African countries’ PEM systems need considerable upgrading, anddonors have identified this as a priority. The World Bank and IMF assessed thePEM systems in Highly Indebted Poor Countries (HIPCs) and found that thesystems in 16 African countries required substantial upgrading, those in 4countries required some upgrading, and only 2 required little upgrading.This suggests that scaling-up scenarios will have to include resources forstrengthening PEM systems.

Over 50 donor agencies, in addition to the World Bank and IMF, areactive in this area and could be called upon for assistance. Instituting well-functioning PEM systems is only one part of the larger effort to consolidatefiscal institutions that many donors are promoting. Policies to introducegreater transparency, strengthen rules governing budget procedures andreporting, and prepare MTEFs are all part of this effort and can provide addi-tional assurance to donors that scaled-up aid will be used effectively(Diamond, 2006). For example, in the Republic of Congo, the recent publi-cation of fiscal data, audited reports on oil activities, and reports on externalverification of government revenues, oil contracts, and data has been a strongsignal to the country’s development partners that a concerted change in thepolicy regime is under way (World Bank and IMF, 2005).

Strengthening Governance 37

Instituting well-functioning PEM systems

is only one part of thelarger effort to

consolidate fiscalinstitutions.


8Maintaining Debt Sustainability

All scaling-up scenarios must be consistent with maintaining public and external debt sustainability. Increased aid can have a significant impacton macroeconomic developments that are fundamental to debt dynamics in the recipient country. Higher aid levels will probably affect the country’s GDP growth rate, fiscal position, interest rates, and balance of payments.

Even if all the additional financing included in the scaling-up scenario isassumed to be in the form of external grants, the debt burden will not neces-sarily improve over time compared with the underlying baseline scenario. Andif part of the scaling up involves higher levels of concessional loan financingor higher domestic borrowing, an updated debt sustainability analysis (DSA),covering both external and total public debt, becomes an absolutely essentialpart of the scaling-up scenario.29

The updated DSA should cover a long time horizon, such as 20 years fromthe base year. This long time frame is necessary because the maturity (and,hence, the debt-servicing implications) of new loans may be long, and princi-pal repayments against the new loans could raise debt-service obligations pre-cisely when the aid inflows are being scaled down to more normal levels.

8.1. Assessing External Debt Dynamics

The most difficult part of conducting a DSA is evaluating the impact ofscaling up on various macroeconomic variables that affect a country’s debtdynamics. Figure 1 depicts the impact of scaling up on external debt. As dis-

38

An updated debtsustainability analysis (DSA)can be an essential part ofthe scaling-up scenario.

29A DSA paper is prepared by the IMF and World Bank staffs and the country authorities. DSAsinclude a comprehensive analysis of the external debt situation of the country concerned and asimulation of the impact of the full use of traditional debt relief mechanisms. DSAs replace HIPCdocuments for countries that have met the requirements to reach a decision point under theHIPC Initiative but which do not require assistance under the Initiative because they have a sus-tainable external debt after full use of traditional debt relief mechanisms.


cussed in other sections of this handbook, scaling up will almost certainlyincrease the country’s imports as aid-financed spending increases. It is alsolikely to change the country’s exports, depending on how aid inflows are usedand how they affect the exchange rate (see Section 2.2). If aid is used toenhance domestic supply capacity and reduce transport costs, for example, itmay increase exports. But if aid leads to appreciation of the real exchangerate, then it could reduce the country’s export prospects compared with thebaseline scenario. This means that, although a real exchange rate apprecia-tion will directly reduce both the country’s foreign debt stock and its debt-ser-vice obligations measured in domestic currency, it may also reduce theavailability of foreign exchange to service debt. On the other hand, a realexchange rate depreciation—for example, as a result of a tapering off of aidor a sudden decline of aid inflows—will increase the burden of foreign debtin domestic currency, although it may boost exports over time.

8.2. Gauging Fiscal Debt Sustainability

Scaling up is also likely to have significant implications for total fiscal(external and domestic) debt through its impact on key macroeconomicvariables (Figure 2). More specifically, scaling up should accelerate GDPgrowth and, hence, generate more government revenue. Scaling up may ormay not change revenue buoyancy (the responsiveness of tax revenue tochanges in income), depending on whether additional policy measures areimplemented. Increased aid may also allow countries whose domestic debtis considered excessive to reduce their domestic debt ratios. A change in thedomestic interest rate as a result of the need to sterilize increased aid inflowswould also raise the cost of servicing domestic debt.

Maintaining Debt Sustainability 39

Figure 1. Impact of Scaling Up on External Debt

Scaling Up

Debt stock and service

Exports

Imports

Exchange rate

Interest rate


8.3. Ensuring Debt Sustainability

If initial projections suggest that a country’s debt will become unsustainablewhen its aid is scaled up, it may seek debt sustainability through a combinationof adjusting the policy framework and, if possible, changing the composition ofaid (using grants instead of loans, or increasing the grant element of new bor-rowing). The type of adjustment required will clearly vary from country to coun-try, depending on existing constraints on domestic policies and the flexibility ofdonor resources. The goal, however, is to provide an appropriate borrowing(and lending) strategy that is consistent with the requirements of scaling up,limits the risk of debt stress, and is supported by a sound macroeconomic frame-work and debt-management institutions.30 A sustainable debt strategy for scal-ing-up countries is to maximize the concessionality of external financing.

The IMF and the World Bank have agreed on a standard framework forconducting a fiscal DSA for scaling up. 31 The key feature of this framework isits policy-dependent, indicative debt-burden thresholds, beyond which debt isconsidered excessive. Countries that have strong domestic policies and insti-tutions are deemed better able to carry external debt than those with weakerpolicies and institutions.32


A sustainable debtstrategy for scaling-upcountries involvesmaximizing theconcessionality of external financing.

30See Daseking and Joshi (2005) for a discussion of the appropriate mix of grants and loans forlow-income countries.

31See IMF and World Bank (2004). A DSA using the new framework was carried out forEthiopia, though not in the context of scaling up (IMF, 2005a).

32The classification of countries by the quality of their policies and institutions is based on theWorld Bank’s Country Policy and Institutional Assessment (CPIA) index. A strong performer(CPIA ≥ 3.75) is considered able to sustain a net present value (NPV) of debt-to-exports ratio of200 percent and an NPV of debt-to-GDP ratio of 50 percent. A poor performer (CPIA ≤ 3.25)is limited to an NPV of debt-to-export ratio of 100 percent and an NPV of debt-to-GDP ratio of30 percent.

Figure 2. Impact of Scaling Up on Fiscal Debt

Scaling Up

Debt stock and service

Revenue

Expenditure

Exchange rate

Interest rate


The external DSA should use multiple indicators. As suggested by theIMF–World Bank framework, these indicators should include three net present value (NPV) ratios for external debt (relative to GDP, exports, andrevenues) and two debt-service ratios (relative to exports and revenues).The NPV debt ratios indicate the future debt obligations a country hastaken on, and the debt-service ratios point to possible liquidity constraintsover time.

It may turn out that a country’s domestic debt is not sustainable even whenits external debt is below the sustainability thresholds. Reiterations of both theexternal and fiscal DSAs may therefore be required to identify the appropriateadjustments to the policy framework and aid composition. Here again, thestandard template can be used to facilitate the DSA after the projections havebeen completed. The following issues should also be considered:

• Public enterprise debt: All public enterprise debt that arises from quasi-fiscal activities and imposes a contingent liability on the governmentshould be included in the analysis, data permitting. In practice, thisexcludes the debt of commercially run public enterprises.33

• Net versus gross domestic debt: It is important to look not only at thedebt a government owes but also at the assets a government has tohelp repay its debt. If, however, government assets are illiquid for thepurpose of debt repayment, the DSA should be conducted on grossdebt.

• Risk of debt distress: The fiscal DSA should use multiple indicators toshow both the government’s capacity to pay its debt obligations and pos-sible liquidity problems. Such indicators may include NPV of debt-to-GDP and debt-to-revenue ratios, foreign-currency-denominated publicdebt as a share of GDP, and the debt-service-to-revenue ratio. Theappropriate mix of these indicators depends on which aspects of thedomestic debt carry the greatest risk for debt distress.

8.4. Strengthening the Debt-Management System

Strengthening debt-management institutions helps reduce the possibilitythat scaled-up lending will cause a country to build up excessive debt. In ascaling-up scenario focused on grants, there may be a temptation to allow theinstitutions associated with debt management to weaken. Debt-managementinstitutions differ from country to country, but their activities generally focuson 1) formulating and communicating debt-management policies and strate-gies, 2) providing the projections and analysis to support policymaking, and3) undertaking operations to implement terms of loan agreements and main-taining comprehensive and up-to-date loan records (Bangura, Kitabire, andPowell, 2000).

Maintaining Debt Sustainability 41

Institutions of debtmanagement must not beallowed to weaken when

aid is scaled up.

33See IMF (2005e).


There are several components of an effective debt-management system:

• All public and publicly guaranteed debt must be contracted under rulesthat are both clearly understood by all public agencies and monitoredcentrally.

• The law should clearly state who is able to contract new borrowing onbehalf of the state and for what purposes. Preferably, all public borrow-ing should be approved by the minister of finance.

• Although private nonguaranteed debt is typically not high in low-incomecountries, it should be monitored by the central government if possiblebecause the servicing of obligations on such debt may have significantmacroeconomic implications. In countries with no exchange controls inplace, data on private debt will need to be obtained through commercialbank records or surveys of private companies.



9Summary: Five Primary Guidelines

Scaling-up scenarios can be used to illustrate the potential impact on acountry of a large increase in external aid and can help identify key mea-sures and policies that will allow the country to absorb and spend increasedaid inflows without destabilizing the macroeconomic environment. Thishandbook stresses that preparing a scaling-up scenario requires a detailedunderstanding of the likely uses of additional aid flows. Ideally, assumptionsabout the uses of aid are based on a detailed assessment by the countryauthorities—done in cooperation with the World Bank, the UN, or otherdonors—of the sectoral needs required to reach the MDGs and the likelypace at which each sector can absorb such funds without encounteringmajor supply bottlenecks.

This handbook identifies five fundamental guidelines for preparing ascaling-up scenario to guide a country’s efforts to achieve the MDGs.

9.1. Minimize the Risks of Dutch Disease

When aid flows increase, a country must choose how much aid to absorband how much to spend. A country may use the opportunity afforded byhigher aid levels to build up its foreign exchange reserves and reduce itsdomestic debt burden, but the typical assumption in scaling-up scenarios isthat the country will both absorb and spend most of the aid. Doing this raises the possibility that the real exchange rate will appreciate, ashigher domestic demand raises the price of nontradables in relation to trad-ables, and that productive resources will be diverted away from the export-ing sectors.

It is unclear how high the risks of such Dutch disease effects are for Africa.The evidence is mixed. In a number of countries, aid surges have been asso-ciated with real exchange rate depreciation, suggesting that supply effectsmore than offset the impact of higher domestic demand on nontradable sec-tors or that aid was not absorbed. When aid is fully absorbed, a real apprecia-tion is likely, at least in the short run.

43


A sound understanding of how a country will spend additional aid, as wellas the likely policy response of the central bank, is therefore critical to pro-jecting the likely macroeconomic impact of a scaling up of aid. Full absorp-tion will put upward pressure on real exchange rates, but there are severalways to mitigate against Dutch disease, including: ensuring that there is a highimport content in additional public spending; focusing spending on infra-structure, which may have a faster impact on productivity; and liberalizingtrade.

9.2. Seek to Enhance Growth in the Short to Medium Term

Governments should aim to implement policies that strengthen thepotential impact of aid on growth. Aid does not automatically increasegrowth, and the impact increasingly seems to depend on the type of aid. Notall aid is directed at activities that will boost growth (for example, emer-gency assistance, humanitarian aid, or disaster relief), and some categoriesof spending only improve growth with long lags (for example, aid for environmental programs, democratic or judicial reform, and some healthand education programs). Aid that might reasonably be expected to affectgrowth in the short to medium term includes aid to build infrastructuresuch as roads, irrigation, ports, and electricity. Hence, there is no simplemethod to estimate the likely growth impact of higher government spending.

There may also be a trade-off between directing aid toward enhancinggrowth (for example, spending on infrastructure) and focusing aid on reliev-ing poverty (for example, aid to rural sectors).

The returns to aid will diminish more quickly in sectors with serious sup-ply bottlenecks, but these are often difficult to identify in advance. It is impor-tant, therefore, for governments to focus on developing policies that will allowthem to effectively absorb higher aid levels and to remain alert to emergingsupply pressures in different sectors. A gradual increase in aid is likely to beabsorbed more easily than a sharp increase.

9.3. Promote Good Governance and Reduce Corruption

Good governance improves the effectiveness of aid in promoting growth.Strengthening institutions for public expenditure management and publicauditing and reducing corruption therefore are likely to increase the benefitsof aid, allowing more funds to be effectively channeled to productive uses andreassuring aid donors that their money is being well spent. Promoting goodgovernance also supports the development bargain emphasized by theCommission for Africa (2005), namely, that country-led reforms should bebacked by increased aid.



9.4. Prepare an Exit Strategy

Governments receiving higher aid flows also face the challenge of what todo if and when the scaled-up aid is not sustained by donors. It may be difficultto reduce expenditures that have been financed by aid, and the pressure forhigher domestic financing of the deficit may increase significantly. Thisunderlines the importance of developing an exit strategy as part of the scal-ing-up scenario, to map the macroeconomic path the country will follow afterscaled-up aid flows fall back to more normal levels or, perhaps to lower-than-normal levels, if aid flows have been front-loaded.

Policies during the scaling-up period must be consistent with a smoothtransition during the scaling-down period. The country must maintain its rev-enue efforts and strengthen its tax systems. The government may choose tosmooth the impact of aid volatility by projecting some accumulation reserves(or decumulation), with associated effects on monetary management.Sufficient current spending must be allowed to ensure that capital investmentcan be operated and maintained.

In a scaling-up scenario that includes significant new loans, it is critical toregularly monitor debt sustainability and to create government institutionsthat can develop and administer a clear and widely understood public debtstrategy. This is important for all low-income countries, whether or not theyhave been granted comprehensive debt relief.

9.5. Regularly Reassess the Appropriate Policy Mix

Scaling-up scenarios are not forecasts. Their precision is conditional onboth sustained higher aid flows and countries’ implementation of policiesthat allow them to absorb the additional aid flows without destabilizing themacroeconomy such as through increased inflation, loss of competitiveness,or a rise of debt to unsustainable levels. Because the scenarios are highlyuncertain, projecting the point at which the returns on aid fall to zero ishighly imprecise. Scenarios should therefore be considered as tools to helpcountries identify the important policy issues they face in using higher aidflows effectively. Countries and donors should use the scenarios to regularlyupdate their vision for the future.

Summary: Five Primary Guidelines 45


Appendix 1. The Relationship between AidFlows and Exchange Rates in

Sub-Saharan Africa

Aid Flows and the Real Exchange Rate

Attempts to measure the relationship between aid flows and the realexchange rate in SSA date back to the early 1980s (Appendix Table A.2.2).

• A number of studies have found a tendency for aid inflows to be associ-ated with an appreciation of the real exchange rate, but this evidence isnot overwhelmingly significant. See Younger (1992) for Ghana, andKasekende and Atingi-Ego (1999) for Uganda, as well as cross-countryanalysis by Adenauer and Vagassky (1998).

• Econometric estimates often show the impact of aid on the exchangerate to be small and statistically insignificant. Bulir and Lane (2002) callthese “traces” of aid-induced real exchange rate appreciation. Prati,Sahay, and Tressel (2003), using a panel data model, suggest that forcountries whose ODA is in excess of 2 percent of GDP a year, a doublingof aid would appreciate the level of real exchange rate by, at most, 4 per-cent in the short run, rising to about 18 percent over a five-year periodand 30 percent over a decade.

• Time-series models tend to reveal that the real exchange rate respondsless to aid variations than to other exogenous factors, such as terms oftrade variations.

• Some studies of African countries find that aid inflows appear to be asso-ciated with a real depreciation, reflecting increased productivity (sup-ply-side response) as a result of aid. See, for example, Nyoni (1998),Sackey (2002), and IMF (2005d). The latter observes that aid that is notabsorbed is not associated with any real exchange rate appreciation,noting that in a number of cases, aid surges went largely into reserves.

Exchange Rates and Exports

Region-specific studies find real exchange rate changes to be a significantdeterminant of the share of exports in GDP in SSA. Balassa (1990) estimates

46


that a 1 percent change in the level of the real exchange rate is associated witha change of 0.8 percent to 1 percent in the share of exports in GDP. Similarly,Ghura and Grennes (1993) find that an actual exchange rate that is 1 percentabove a model-based equilibrium exchange rate lowers the share of exportsby 0.096 percent. Rajan and Subramanian (2005a) argue that in countriesthat receive more aid, export-oriented, labor-intensive industries grow moreslowly than other industries, suggesting that aid does lead to Dutch disease.

However, the impact of exchange rate instability on exports can also be animportant consideration. With a smaller sample of countries, Sekkat andVaroudakis (2000) estimate a higher elasticity between exchange rate mis-alignment and the share of exports in some sectors, but this result is notalways significant. These studies, however, do not estimate the real exchangerate effects in the context of other factors that might hinder export growth.

To the extent that higher aid flows alleviate supply bottlenecks, they canoffset the effect of an exchange rate appreciation on export growth. In a studycovering 60 developing countries, Elbadawi (2002) finds that real exchangerate depreciation did play a significant role in export growth (with the elas-ticity varying between 0.54 and 0.64 for the entire group). But the coefficientsof regional dummies in the study suggest that all other things equal, nontra-ditional exports from East Asia and Latin America would be higher than thosefrom SSA, implying that supply constraints, not included in the model, mightsignificantly account for the poor performance of nontraditional exports inthis region.

Exchange Rates and Economic Growth

Recent studies find that growth accelerations are associated with realdepreciation, suggesting that a large real appreciation associated with scalingup could have long-term growth costs. Several studies have built onHausmann, Pritchett, and Rodrik’s (2004) analysis of jumps in countries’medium-term growth trends, which they label “growth accelerations.” Theirstudy found that the onset of accelerations had a strong correlation with realexchange rate depreciation. This finding has been confirmed for SSA (IMF,2005f). Almost all the sustained growth cases in SSA avoided overvaluationduring the growth period. The study also notes the close link between avoid-ance of exchange rate misalignment and macroeconomic stability, reinforc-ing the case for aid inflows to be accompanied by prudent macroeconomicmanagement.

Overvaluation of the exchange rate dampens growth. With a dataset thatincludes 73 developing countries, spanning the period from 1975 to 1992,Razin and Collins (1997) find that overvaluation does have a significant neg-ative impact on growth, while there is no statistically significant relationshipbetween undervaluation and growth. According to their estimates, a 1 percentovervaluation is associated with a 0.06 percent decline in the real per capitagrowth rate. Dollar (1992) estimates that a 1 percent distortion of theexchange rate dampened the per capita growth rate by about 0.02 percent for

Appendix 1 47


95 developing countries between 1976 and 1985.34 Similarly, Cottani, Cavallo,and Khan (1990) estimate that for a group of 24 developing countries overthe period 1960–83, the growth-dampening effect of exchange rate misalign-ment was about 0.1 percent. Ghura and Grennes (1993) find that a 1 percentovervaluation dampens real per capita GDP growth by about 0.02 percent.Bleaney and Greenway (2001) also estimate a negative effect of laggedexchange rate misalignment on growth. They estimate, with data covering 14SSA countries over the period 1980–95, that a 1 percent lagged misalignmentdampens GDP growth by 0.04 percent.

Exchange Rates and Investment

One of the channels through which a temporarily stronger exchange ratemay influence the growth rate is the impact on investment. Razin and Collins(1997) posit that, in addition to its effect on the competitiveness of the trad-ables sector, a stronger exchange rate may also affect domestic and foreigninvestment, thereby influencing the capital accumulation process. Bleaneyand Greenway (2001) suggest that an overvaluation might hurt investmenteven though it lowers the price of imported capital goods, because it reducesthe returns to investment in the tradables sector and because the resultingcurrent account deficit creates the need for tighter macroeconomic policies.


34In this case, the right-hand-side variable is not exchange rate misalignment but exchange ratedistortion. Whereas misalignment indicates the extent to which an exchange rate is overvaluedgiven the fundamentals, Dollar’s index specifically measures the extent to which the exchangerate is distorted away from its free-trade level.


Appendix 2. The Macroeconomics of Aid

49


50M

AC

ROEC

ON

OM

ICC

HA

LLENG

ESO

FSC

ALIN

GU

PA

IDTO

AFRIC

A

Appendix Table A.2.1 Official Net Transfers to SSA Countries(In percent of GDP)

1980–84 1985–89 1990–94 1995–99 2000–03 1980–2003

(More than 15 percent)

Equatorial Guinea 47 São Tomé and Príncipe 38 São Tomé and Príncipe 78 São Tomé and Príncipe 52 Eritrea 34 São Tomé and Príncipe 46Cape Verde 39 Guinea-Bissau 35 Mozambique 36 Guinea-Bissau 24 São Tomé and Príncipe 33 Guinea-Bissau 30Guinea-Bissau 36 Cape Verde 30 Guinea-Bissau 29 Mozambique 23 Mozambique 29 Mozambique 27Comoros 29 Equatorial Guinea 29 Eritrea 26 Rwanda 23 Sierra Leone 28 Eritrea 25São Tomé and Príncipe 26 Gambia, The 26 Cape Verde 21 Liberia 20 Congo, Dem. Rep. of 25 Cape Verde 23Mozambique 22 Mozambique 22 Equatorial Guinea 20 Malawi 18 Guinea-Bissau 25 Equatorial Guinea 21Gambia, The 18 Comoros 20 Rwanda 20 Eritrea 18 Burundi 22 Comoros 16Lesotho 16 Lesotho 17 Burundi 20 Cape Verde 15 Malawi 17 Gambia, The 16Ethiopia 15 Malawi 19 Malawi 15

Gambia, The 17Uganda 17Tanzania 16Zambia 15

(Between 5 and 15 percent)

Mali 12 Malawi 14 Chad 13 Sierra Leone 12 Ethiopia 14 Burundi 15Senegal 11 Chad 14 Niger 13 Burundi 11 Rwanda 13 Liberia 14Burundi 11 Mali 13 Lesotho 13 Niger 10 Uganda 11 Rwanda 14Seychelles 10 Burundi 12 Comoros 13 Chad 10 Niger 11 Ethiopia 12Central African Rep. 9 Niger 12 Sierra Leone 12 Central African Rep. 10 Zambia 11 Sierra Leone 12Tanzania 9 Zambia 12 Ethiopia 12 Burkina Faso 10 Tanzania 11 Lesotho 11Malawi 9 Ethiopia 11 Central African Rep. 12 Zambia 10 Liberia 10 Mali 11Burkina Faso 9 Tanzania 10 Mali 11 Comoros 10 Burkina Faso 10 Tanzania 11Chad 8 Central African Rep. 10 Benin 10 Madagascar 9 Gambia, The 9 Niger 11Niger 7 Burkina Faso 9 Madagascar 10 Mali 9 Mali 9 Zambia 11Botswana 6 Benin 8 Burkina Faso 10 Ethiopia 8 Cape Verde 9 Chad 11Benin 6 Senegal 8 Guinea 9 Uganda 8 Ghana 8 Burkina Faso 9Madagascar 6 Guinea 8 Ghana 9 Tanzania 8 Chad 7 Central African Rep. 9


Appendix 251

Rwanda 6 Madagascar 7 Senegal 9 Benin 7 Comoros 7 Uganda 8Zambia 6 Togo 7 Togo 7 Senegal 7 Benin 6 Madagascar 8Congo, Rep. of 6 Rwanda 6 Kenya 6 Gambia, The 7 Madagascar 6 Senegal 8Kenya 5 Seychelles 6 Côte d’Ivoire 6 Ghana 7 Benin 8

Ghana 6 Zimbabwe 5 Togo 6 Congo, Dem. Rep. 6Sierra Leone 6 Guinea 5 Ghana 6Kenya 5 Lesotho 5 Togo 6

Guinea 6

(Less than 5 percent)

Swaziland 4 Congo, Rep. of 5 Cameroon 4 Equatorial Guinea 5 Central African Rep. 5 Seychelles 5Togo 4 Uganda 5 Congo, Dem. Rep. of 4 Congo, Rep. of 3 Senegal 4 Kenya 4Zimbabwe 4 Botswana 5 Angola 4 Angola 2 Lesotho 4 Congo, Rep. of 4Mauritius 4 Congo, Dem. Rep. of 3 Congo, Rep. of 4 Congo, Dem. Rep. 2 Togo 3 Zimbabwe 3Sierra Leone 3 Zimbabwe 3 Seychelles 3 Seychelles 2 Seychelles 3 Angola 2Côte d’Ivoire 2 Mauritius 2 Gabon 1 Zimbabwe 2 Guinea 3 Botswana 2Uganda 2 Angola 2 Botswana 1 Swaziland 1 Cameroon 2 Côte d’Ivoire 2Cameroon 2 Gabon 2 Swaziland 0 Kenya 1 Kenya 2 Cameroon 2Guinea 2 Cameroon 2 South Africa 0 Côte d’Ivoire 1 Zimbabwe 1 Swaziland 1Congo, Dem. Rep. of 1 Côte d’Ivoire 1 Mauritius 0 Cameroon 0 Equatorial Guinea 1 Mauritius 1Ghana 1 Swaziland 0 Nigeria –2 South Africa 0 Angola 1 South Africa 0Gabon 1 Nigeria 0 Liberia n.a. Botswana 0 Swaziland 1 Gabon –1Nigeria 0 Liberia n.a. Namibia n.a. Mauritius –1 Côte d’Ivoire 1 Nigeria –1South Africa n.a. South Africa n.a. Nigeria –2 Congo, Rep. of 1 Namibia n.a.Eritrea n.a. Namibia n.a. Gabon –3 South Africa 0Namibia n.a. Eritrea n.a. Namibia n.a. Mauritius 0Angola n.a. Botswana 0Liberia n.a. Nigeria –2

Gabon –5Namibia n.a.

Source: World Bank (2005).Note: n.a. denotes not available.


52M

AC

ROEC

ON

OM

ICC

HA

LLENG

ESO

FSC

ALIN

GU

PA

IDTO

AFRIC

A

Table A.2.2. Aid Inflows and Real Exchange Rate Appreciation: Empirical Evidence

Aid and RealExchange Rate Appreciation

Study Sample Relationship Effect of 1 Percent Real Increase in Aid

Cross-Sectional Studies

Van Wijnbergen (1985) 6 African countries, 1969–83 Mixed Appreciation of 0.2–0.9 percent over two years for some countries;no significant change for others

Adenauer and Vagassky (1998) 4 CFA franc zone countries, 1980–92 Positive Appreciation of 0.13 percent over 2 years

Bulir and Lane (2002) 9 developing countries Positive n.a.

Prati, Sahay, and Tressel (2003) 87 developing countries, 1960–98 Positive Appreciation of 0.04 percent in the first year

Elbadawi (1999) 62 developing countries (28 from Positive Appreciation of 0.09 percent in the first year1

Africa), 1984–85, 1989–90, 1994–95

Yano and Nugent (1999) 44 developing countries, 1970–90 Mixed n.a.

Country-Specific Studies

Younger (1992) Ghana, 1960–88 Positive n.a.

Kasekende and Atingi-Ego (1999) Uganda, 1970–96 Positive Appreciation of 0.03 percent in the first year

Nyoni (1998) Tanzania, 1967–93 Negative Depreciation of 0.13 percent in the first year

Sackey (2002) Ghana, 1962–96 Negative Depreciation of 0.03 percent in the first year

IMF (2005e) Ethiopia, Ghana, Mozambique, Negative Depreciation of 1.5–6.5 percent the year following an aid surge2

Tanzania, Uganda

Sources: As cited

n.a. indicates that the study does not yield an elasticity measure.1Indicates response to 1 percent increase in ratio of ODA to GNP.2Indicates response to an aid surge with inflows ranging between 2 percent and 11 percent of GDP.


Appendix 253

Table A.2.3. Empirical Studies of Kinks in the Relationship between Inflation and Growth

Growth effect of Inflation threshold higher inflation

(percent) below the threshold Countries Period Inflation measure Remarks

Fischer (1993) 15 Negative 80 1960–89 CPI

Barro (1991) 10–20 Not significant 117 1960–90 10-year average CPI

Sarel (1996) 8 Positive 87 1970–90 5-year average CPI

Bruno and Easterly (1998) 40 Not significant 97 1961–92 CPI

Ghosh and Phillips (1998) >5 Positive 145 1960–96 Average annual CPI

Kochhar and Coorey (1999) 5 Positive 84 1981–95 Average annual CPI(low- and middle-income countries only)

Khan and Senhadji (2000) 7–11 (for developing Positive 140 1960–98 5-year average CPI 1–3 percent threshold forcountries only) industrial countries

Controlled for investmentand unemployment

Burdekin, Denzou, Keil, Sitthiyot, and Willett (2000) 3 (for developing Positive 51 1967–92 8 percent threshold for

countries only) industrial countries

Gylfason and Herbertsson 10–20 Positive 170 1960–92 5-year average GDP (2001) deflator

CPI = consumer price index.



Table A.2.4. Incremental Effect of Aid on Domestic Revenue

Study Sample Domestic Revenue1

Heller (1975) 9 Anglophone African countries, 1960–71 –0.42

Pack and Pack (1990) Indonesia, 1966–86 0.29

Cashel-Cordo and Craig (1990) 46 Least Developed (African countries) 10.29 Countries, 1975–80 (non-African countries) –4.25

Gang and Khan (1990) India, 1961–84 0.00

Khilji and Zampelli (1991) Pakistan, 1960–86 -0.01

Leuthold (1991) 8 African countries, 0.001973–81

Khan and Hoshino (1992) 5 Asian countries, 1.201955–76

Gupta (1993) India, 1969–93 0.01

Pack and Pack (1993) Dominican Republic, –0.391968–86

Rubino (1997) Indonesia –1.40

Iqbal (1997) Pakistan, 1976–95 0.00

Franco-Rodrigues, Morrissey, Pakistan, 1960–95 (Direct effects) –2.90and McGillivray (1998)

(Total effects) –3.60

MacGillivray and Ahmed (1999) The Philippines, 0.101960–92

Franco-Rodriguez (2000) Costa Rica, 1971–94 1.10

McGillivray (2000) Pakistan, 1956–95 0.00

Swaroop, Jha, and Sunil Rajkumar (2000) India, 1970–95 0.00

Sources: McGillivray and Morrisey (2001); Feeny and McGillivray (2003); as cited.1Figures are the total effect of a one unit increment in aid on domestic revenue collections.


Appendix 2 55

Table A.2.5. Tax Revenue and Trade Taxes, by Region(In percent of GDP)

International Trade Taxes_______________________________Tax Revenue Import duties Export duties______________ ______________ ______________

Early Early Early Early Early EarlyCountry Subgroups1 1990s 2000s 1990s 2000s 1990s 2000s

Americas 14.9 16.3 2.5 1.9 0.2 0.0Sub-Saharan Africa 16.3 15.9 4.9 3.5 1.0 0.4Central Europe and BRO2 27.3 23.4 1.4 0.9 0.8 0.4North Africa and Middle East 15.1 17.1 3.6 3.0 0.1 0.1Asia and Pacific 13.6 13.2 3.2 1.9 0.3 0.2Small islands 25.5 24.5 13.5 9.7 0.3 0.0

Unweighted averageDeveloping countries3 17.9 17.6 3.9 2.7 0.5 0.2High-income countries 26.6 27.5 2.0 1.3 0.0 0.2PRGF–eligible countries 15.2 14.8 4.8 3.5 0.6 0.3

Source: Keen and Simone (2004).1Subgroups contain only developing countries.2Baltic countries, Russia, and the other countries of the former Soviet Union.3Defined as low- and middle-income countries.

Table A.2.6. Relationship between Public and Private Investment in Sub-Saharan Africa

Study Data and Coverage Results

Greene and Villanueva (1991) 23 developing countries Public investment increases (Asia, SSA, and Latin private investmentAmerica); pooled sample

Hadjimichael and Ghura 41 SSA countries; panel Public investment increases (1995) data private investment; also

studied other policydeterminants of privateinvestment

Odedokun (1997) 48 developing countries Public infrastructure (SSA, Asia, Europe, and investment increases North Africa); panel data private investment; non-

infrastructure crowds outprivate investment

Ghura and Goodwin (2000) 31 developing countries Public investment increases (Asia, SSA, and Latin private investment in SSA, America); panel data but crowds out in Asia and

Latin America

Belloc and Vertova (2004) 7 HIPC countries, 5 in Public investment increases SSA; country-level vector private investment in 6 of autoregressions 7 countries



Table A.2.7. The Effect of Public Investment on Output


Canning and Bennathan Penn World Tables and Output elasticity with respect (2000) specialized infrastructure to roads and electricity in

data; annual; all country range of 0.05–1.0; highest income levels for MICs; strong input

complementarity

Calderon and Serven World Bank data with Large output gains to closing (2004) infrastructure indices; infrastructure gaps

5-year averages; all country income levels

Barro (1991) Penn World Tables, World No impact of public Bank and UN; averages, investment on growth1960–85; all country income levels

Easterly and Rebelo Specialized data capturing No impact of public (1993) broad public investment; investment aggregate on per

10-year averages; all capita growth, but big impact country income levels (0.6) for transport and

communication and directpublic investment of thegovernment (0.4)

Devarajan, Swaroop, and Government Financial Studies impact of changes in Zou (1996) Statistics; 5-year averages; budget composition; capital

LICs and MICs spending has negative effect

Khan and Kumar (1997) Penn World Tables, World Public investment has positive Bank, World Economic impact on per capita growth Outlook; 10-year averages; (0.3), but weakening in recent LICs and MICs samples

Clements, Bhattacharya, World Bank; 3-year Public investment has positive and Nguyen (2003) averages; LICs impact (0.2) when not deficit-

financed

Gupta, Clements, Baldacci, IMF; annual; LICs Positive impact of public and Mulas-Granados capital outlays (0.7) on per (2004) capita growth

Source: Adapted from Table 1 in IMF (2004a) to those studies that include evidence from LICs.Note: Growth regression studies use country average growth rates typically calculated over

5- or 10-year periods. The numbers in parentheses in the third column indicate the estimatedresponse of this average to a permanent increase in the expenditure share of 1 percent of GDP.They should thus be interpreted as the long-run response to a small permanent change in theexpenditure share. While Gupta Clements, Baldacci, and Mulas-Granados (2004) use annualgrowth data, their specification allows them to distinguish short- and long-run impacts.

LIC = low-income country; MIC = middle-income country.


Appendix 2 57

Table A.2.8. Impact of Health and Education Sectors on Output


Barro (1991) Penn World Tables, World Significant role for human Bank and UN; averages capital (especially initial stock 1960–85; all country of secondary education)income levels

Barro and Sala-i-Martin Data similar to Barro’s Positive impact of education (1995) (1991); 10-year averages; spending on per capita

all country income levels growth (0.2)

Devarajan, Swaroop, and Government Financial Impact of changes in budget Zou (1996) Statistics; 5-year averages; composition; current spending

LICs and MICs has positive impact, mixedresults for functionalbreakdown

Krueger and Lindahl Similar to Barro (1991) Big impact of change in (2001) augmented with World schooling on growth, but only

Values Survey; data detectable in long-period averages of various lengths; averages (10-20 years); size of all country income levels effect varies with econometric

specification

Baldacci and others World Bank and IMF; Positive impact of spending (2004) 5-year averages; LICs and on education (0.5) and health

MICs (0.4) after 5 years; educationimpact rises to 1.4 after 15years; diminishing returns tolevel of education and healthspending

Canning, and Sevilla Penn World Tables, UN, Studies impact of health Bloom (2003) and World Bank; 10-year indicators and not

averages; all country expenditure; 1-year increase in income levels average life expectancy raises

output by 4 percent

Note: These studies use country average growth rates typically calculated over 5- or 10-yearperiods. The numbers in parentheses in the third column indicate the estimated response ofthis average to a permanent increase in the expenditure share of 1 percent of GDP. They shouldthus be interpreted as the long-run response to a small permanent change in the expenditureshare.

LIC = low-income country; MIC = middle-income country.



Table A.2.9. Impact of Increasing Corruption by One Unit1

Study Impact on Finding

Mauro (1995) Real per capita GDP growth –0.3 to –1.8 percentage points

Leite and Weidmann Real per capita GDP growth –0.7 to –1.2 percentage points(2002)

Tanzi and Davoodi Real per capita GDP growth –0.6 percentage points(2002)

Abed and Davoodi Real per capita GDP growth –1 to –1.3 percentage points(2002)

Mauro (1995) Ratio of investment to GDP –1 to –2.8 percentage points

Mauro (1998) Ratio of public education –0.7 to –0.9 percentage pointsspending to GDP

Mauro (1998) Ratio of public health –0.6 to –0.9 percentage pointsspending to GDP

Gupta, Davoodi, and Income inequality (Gini +3.5 to +4.25 Gini pointsAlonso-Terme (2002) coefficient)

Gupta, Davoodi, and Income growth of the poor –2 to –10 percentage pointsAlonso-Terme (2002)

Ghura (2002) Ratio of tax revenues to GDP –1 to –2.9 percentage points

Tanzi and Davoodi Measures of government –0.1 to –2.7 percentage points(2002) revenues to GDP ratio

Gupta, de Mello and Ratio of military spending to + 1 percentage pointSharan (2002) GDP

Gupta, Davoodi, and Child mortality rate + 1.1 to 1.5 deaths per 1,000 Tiongson (2002) live births

Gupta, Davoodi, and Primary student dropout + 1.1 to 1.4 percentage pointsTiongson (2002) rate

Tanzi and Davoodi Ratio of public investment to + 0.5 percentage points(1998) GDP

Tanzi and Davoodi Percent of paved roads in –2.2 to –3.9 percentage points(1998) good condition

Sources: IMF, Fiscal Affairs Department; Transparency International (2001).1Corruption is measured on a scale of 0 (highly clean) to 10 (highly corrupt).Most of the above studies use the Transparency International corruption measure, rescaled

so that higher values in the range 1-10 correspond to higher corruption. When theInternational Country Risk Guide measure of corruption was used, it was rescaled in the sameway. Both the Transparency International and International Country Risk Guide measures relyon expert perceptions. For Gupta, Davoodi, and Tiongson (2000), an index was constructedfrom National Service Delivery Surveys. This has the advantage of being based on reported ser-vice client experience.


References

Abed, George T., and Hamid R. Davoodi, 2002, “Corruption, Structural Reforms, andEconomic Performance in the Transition Economies,” in Governance, Corruption,and Economic Performance, ed. by George T. Abed and Sanjeev Gupta (Washington:International Monetary Fund), chapter 18.

Adam, Christopher, 2004, “Fiscal Policy Design in Low-Income Countries,” in FiscalPolicy for Development: Poverty, Reconstruction, and Growth, ed. by Tony Addison andAlan Roe (New York: Palgrave MacMillan).

———, 2005, “Exogenous Inflows and Real Exchange Rates: Theoretical Quirk orEmpirical Reality?” paper presented at IMF Seminar on Foreign Aid andMacroeconomic Management, Maputo, Mozambique, March 14–15. Available at:http://www.imf.org/external/np/seminars/eng/2005/famm/pdf/adam.pdf

Adam, Christopher S., and David L. Bevan, 2003, “Aid, Public Expenditure and DutchDisease,” CSAE Working Paper 184 (Oxford: Centre for the Study of AfricanEconomies, Oxford University). Available at: http://www.bepress.com/csae/paper184

———, 2004, “Aid and the Supply Side: Public Investment, Export Performance andDutch Disease in Low Income Countries,” Department of Economics DiscussionPaper Series 201 (Oxford: Department of Economics, Oxford University).Available at: http://www.economics.ox.ac.uk/Research/wp/pdf/paper201.pdf

Adenauer, Isabell, and Laurence Vagassky, 1998, “Aid and the Real Exchange Rate:Dutch Disease Effects in African Countries,” Intereconomics, Vol. 33, No. 4, pp. 177–85.

Agbeyegbe, Terence D., Janet Gale Stotsky, and Asegedech WoldeMariam, 2004,“Trade Liberalization, Exchange Rate Changes, and Tax Revenue in Sub-Saharan Africa,” IMF Working Paper 04/178 (Washington: International MonetaryFund).

Agence Française de Développement, Bundesministerium für WirtschaftlicheZusammenarbeit und Entwicklung, Kreditanstalt für Wiederaufbau Entwicklungs-bank, U.K. Department for International Development, and the World Bank, 2005.“Pro-Poor Growth in the 1990s: Lessons and Insights from 14 Countries,” WorkingPaper 32885 (Washington: World Bank).

Andrews, David, Lodewyk Erasmus, and Robert Powell, 2005, “Ethiopia: Scaling Up,”Finance and Development, Vol. 42, No. 3 (September), pp. 32–35.

Atingi-Ego, Michael, 2005. “Choice of Appropriate Monetary Policy Instrument Mix byBank of Uganda” (unpublished; Kampala: Bank of Uganda).

Balassa, Bela A., 1990, “Incentive Policies and Export Performance in Sub-SaharanAfrica,” World Development, Vol. 18, No. 3 (March), pp. 383–91.

Baldacci, Emanuele, Benedict J. Clements, Sanjeev Gupta, and Qiang Cui, 2004.“Social Spending, Human Capital, and Growth in Developing Countries:Implications for Achieving the MDGs,” IMF Working Paper 04/217 (Washington:International Monetary Fund).

59


http://www.imf.org/external/np/seminars/eng/2005/famm/pdf/adam.pdf

http://www.bepress.com/csae/paper184

http://www.economics.ox.ac.uk/Research/wp/pdf/paper201.pdf

Baldacci, Emanuele, and Kevin Fletcher, 2004, “A Framework for Fiscal DebtSustainability Analysis in Low-Income Countries,” in Helping Countries Develop: TheRole of Fiscal Policy, ed. by Sanjeev Gupta, Gabriela Inchauste, and BenedictClements (Washington: International Monetary Fund), chapter 6.

Bangura, Sheku, Damoni N. Kitabire, and Robert Powell, 2000, “External DebtManagement in Low-Income Countries,” IMF Working Paper 00/196 (Washington:International Monetary Fund).

Barro, Robert J., 1991, “Economic Growth in a Cross Section of Countries,” QuarterlyJournal of Economics, Vol. 106, No. 2 (May), pp. 407–44.

———, and Xavier Sala-i-Martin, 1995, Economic Growth (New York: McGraw-Hill).

Bauer, Peter T., 1972, Dissent on Development (Cambridge: Harvard University Press).

Belloc, Marianna, and Pietro Vertova, 2004. “How Does Public Investment AffectEconomic Growth in HIPC? An Empirical Assessment,” Working Paper No. 416(Siena, Italy: University of Siena, Department of Policitical Economics). Availableat: http://www.econ-pol.unisi.it/quaderni/416.pdf

Berg, Andy, 2005, “High Aid Inflows Case Study: Ghana,” paper presented at IMFSeminar on Foreign Aid and Macroeconomic Management, Maputo,Mozambique, March 14–15. Available at: http://www.imf.org/external/np/seminars/eng/2005/famm/pdf/berg.pdf

Bevan, David L., 2005. “An Analytical Overview of Aid Absorption: Recognizing andAvoiding Macroeconomic Hazards,” paper presented at IMF Seminar on ForeignAid and Macroeconomic Management, Maputo, Mozambique, March 14–15.Available at: http://www.imf.org/external/np/seminars/eng/2005/famm/pdf/bevan.pdf

Bleaney, Michael, and David Greenaway, 2001, “The Impact of Short Terms of Tradeand Real Exchange Rate Volatility on Investment and Growth in Sub-SaharanAfrica,” Journal of Development Economics, Vol. 65, No. 2 (August), pp. 491–500.

Bloom, David E., David Canning, and Jaypee Sevilla, 2004. “The Effect of Health onEconomic Growth: A Production Function Approach,” World Development, Vol. 32,No. 1 (January), pp. 1–13.

Boone, Peter, 1996, “Politics and the Effectiveness of Foreign Aid,” European EconomicReview, Vol. 40, No. 2, pp. 289–330.

Briceño-Garmendia, Cecilia, Antonio Estache, and Nemat Shafik, 2004,“Infrastructure Services in Developing Countries: Access, Quality, Costs and PolicyReform,” Policy Research Working Paper No. 3468 (Washington: World Bank).

Bruno, Michael, and William Easterly, 1998, “Inflation Crises and Long Run Growth,”Journal of Monetary Economics, Vol. 41, No. 1, pp. 3–26.

Bulir, Ales , and A. Javier Hamann, 2005, “Volatility of Development Aid: From theFrying Pan into the Fire?” paper presented at IMF Seminar on Foreign Aid andMacroeconomic Management, Maputo, Mozambique, March 14–15. Available at:http://www.imf.org/external/np/seminars/eng/2005/famm/pdf/hamann.pdf

Bulir, Ales , and Timothy Lane, 2002, “Aid and Fiscal Management,” IMF WorkingPaper 02/112 (Washington: International Monetary Fund).

Burdekin, Richard C.K., Arthur T. Denzau, Manfred W. Keil, Thitithep Sitthiyot, andThomas D. Willett, 2000, “When Does Inflation Hurt Economic Growth? DifferentNonlinearities for Different Economies,” Working Paper 22 (Claremont,California: Claremont Colleges).

Burnside, Craig, and David Dollar, 2000,”Aid, Policies and Growth,” American EconomicReview, Vol. 90, No. 4 (September), pp. 847–68.

Calderon, Cesar, and Luis Serven, 2004, “The Effects of Infrastructure Developmenton Growth and Income Distribution,” Policy Research Working Paper No. 3400(Washington: World Bank).



http://www.economics.ox.ac.uk/quaderni/416.pdf

http://www.imf.org/external/np/seminars/eng/2005/famm/pdf/berg.pdf

http://www.imf.org/external/np/seminars/eng/2005/famm/pdf/bevan.pdf

http://www.imf.org/external/np/seminars/eng/2005/famm/pdf/hamann.pdf

http://www.imf.org/external/np/seminars/eng/2005/famm/pdf/berg.pdf

Canning, David, and Esra Bennathan, 2000, “The Social Rate of Return onInfrastructure Investments,” Policy Research Working Paper No. 2390 (Washington:World Bank).

Canning, David, David Bloom, and Jaypee Sevilla, 2003, “Geography and PovertyTraps,” Journal of Economic Growth, Vol. 8, pp. 355–78.

Cashel-Cordo, Peter, and Steven Craig 1990, “The Public Sector Impact ofInternational Resource Transfers,” Journal of Development Economics, Vol. 32, No. 1(January), pp. 17–42.

Chauvet, Lisa, and Paul Collier, 2004, “Development Effectiveness in Fragile States:Spillovers and Turnarounds,” CSAE Working Paper (Oxford: Centre for the Studyof African Economies, Oxford University).

Chauvet, Lisa, and Patrick Guillaumont, 2004, “Aid and Growth Revisited: Policy,Economic Vulnerability and Political Instability,” in Toward Pro-Poor Policies Aid,Institutions, and Globalization, ed. by Bertil Tungodden, Nicholas Stern, and IvarKolstad (New York: Oxford University Press and World Bank).

Chenery, Hollis B., and Alan M. Strout, 1966, “Foreign Assistance and EconomicDevelopment,” American Economic Review, Vol. 56, No. 4, pp. 679–733.

Chu, Ke-young, Sanjeev Gupta, Jerald Alan Schiff, and others, 1995, Unproductive PublicExpenditures: A Pragmatic Approach to Policy Analysis, IMF Pamphlet Series, No. 48(Washington: International Monetary Fund).

Clemens, Michael, and Steven Radelet, 2003, “Absorptive Capacity: How Much Is TooMuch?” in Challenging Foreign Aid: A Policymaker’s Guide to the Millennium ChallengeAccount, ed. by Steven Radelet (Washington: Center for Global Development),chapter 7.

———, and Rikhil Bhavnani, 2004, “Counting Chickens When They Hatch: The Short-term Effect of Aid on Growth,” Working Paper 44 (Washington: Center for GlobalDevelopment).

Clements, Benedict J., Rina Bhattacharya, and Toan Quoc Nguyen, 2003, “ExternalDebt, Public Investment, and Growth in Low-Income Countries,” IMF WorkingPaper 03/249 (Washington: International Monetary Fund).

Collier, Paul, and Jan Dehn, 2001, “Aid, Shocks, and Growth,” Policy ResearchWorking Paper No. 2688 (Washington: World Bank).

Collier, Paul, and David Dollar, 2002, “Aid Allocation and Poverty Reduction,”European Economic Review, Vol. 46, No. 8 (September), pp. 1475–1500.

Collier, Paul, and Anke Hoeffler, 2004, “Aid, Policy and Growth in Post-ConflictSocieties,” European Economic Review, Vol. 48, No. 5 (October), pp. 1125–45.

Collier, Paul, and Steve O’Connell, 2005, “Opportunities, Choices and Syndromes,”paper presented at AERC/Harvard workshop on Explaining African EconomicGrowth, 1960–2000, Weatherhead Center for International Affairs, HarvardUniversity, Cambridge, Massachusetts, March 18–19, 2005. Available at: http://www.wcfia.harvard.edu/conferences/batesafrica/PapersPDF/Collier O’C onnell.pdf

Commission for Africa, 2005, Our Common Interest: Report of the Commission for Africa(London: Department for International Development). Available at:http://www.commissionforafrica.org/english/report/thereport/cfafullreport_1.pdf

Cottani, Joaquin A., Domingo F. Cavallo, and M. Shahbaz Khan, 1990, “Real ExchangeRate Behavior and Economic Performance in LDCs,” Economic Development andCultural Change, Vol. 39, No. 1 (October), pp. 61–72.

Daseking, Christina, and Bikas Joshi, 2005, “Debt and New Financing in Low-IncomeCountries, “ paper presented at IMF Seminar on Foreign Aid and MacroeconomicManagement, Maputo, Mozambique, March 14–15. Available at: http://www.imf.org/external/np/seminars/eng/2005/famm/pdf/bikas.pdf.

References 61


http://www.wcfia.harvard.edu/conferences/batesafrica/PapersPDF/Collier%20O%E2%80%99C%20onnell.pdf

http://www.commissionforafrica.org/english/report/thereport/cfafullreport_1.pdf

http://www.imf.org/external/np/seminars/eng/2005/famm/pdf/bikas.pdf

http://www.wcfia.harvard.edu/conferences/batesafrica/PapersPDF/Collier%20O%E2%80%99C%20onnell.pdf

http://www.imf.org/external/np/seminars/eng/2005/famm/pdf/bikas.pdf

Davoodi, Hamid R., Erwin R. Tiongson, and Sawitree S. Asawanuchit, 2003, “HowUseful Are Benefit Incidence Analyses of Public Education and Health Spending?”IMF Working Paper 03/227 (Washington: International Monetary Fund).

de Mariz Rozeira, Christine, 2005, “Tax Treatments Applicable to Foreign-FinancedProjects: Experience and Principles”(unpublished; Washington: InternationalMonetary Fund).

de Renzio, Paolo, 2004, “The Challenge of Absorptive Capacity,” report on a seminarheld at U.K. Department for International Development, London, September 9(London: Overseas Development Institute). Available at: http://www.odi.org.uk/pppg/cape/Absorptive%20Capacity%20Seminar%20-%20Final%20Report.pdf.

———, 2005, “Scaling Up Versus Absorptive Capacity: Challenges and Opportunitiesfor Reaching the MDGs in Africa,” ODI Briefing Paper (London: OverseasDevelopment Institute).

Devarajan, Shantayanan, Vinaya Swaroop, and Hengfu Zou, 1996, “The Compositionof Public Expenditure and Economic Growth,” Journal of Monetary Economics, Vol.37, No. 2, pp. 313–44.

Diamond, Jack, 2006, Budget System Reform in Emerging Economies: The Challenges and theReform Agenda, Occasional Paper No. 245 (forthcoming; Washington: InternationalMonetary Fund).

Dollar, David, 1992, “Outward-Oriented Developing Economies Really Do Grow MoreRapidly: Evidence from 95 LDCs, 1976–1985,” Economic Development and CulturalChange, Vol. 40, No. 3, pp. 523–44.

Easterly, William, 2001, The Elusive Quest for Growth: Economists’ Adventures andMisadventures in the Tropics (Cambridge, Massachusetts: MIT Press).

———, 2003, “Can Foreign Aid Buy Growth?” Journal of Economic Perspectives, Vol. 17,No. 3, pp. 23–48.

———, and Sergio Rebelo, 1993, “Fiscal Policy and Economic Growth: An EmpiricalInvestigation,” Journal of Monetary Economics, Vol. 32, No. 3, pp. 417–58.

Eifert, Benn, and Alan Gelb, 2005, “Improving the Dynamics of Aid: Towards MorePredictable Budget Support,” Policy Research Working Paper No. 3732(Washington: World Bank).

Elbadawi, Ibrahim, 1999, “External Aid: Help or Hindrance to Export Orientation inAfrica?” Journal of African Economies, Vol. 8, pp. 578–616.

———, 2002, “Real Exchange Rate Policy and Non-Traditional Exports in DevelopingCountries,” in Non-Traditional Export Promotion in Africa: Experience and Issues, ed. byGerald K. Helleiner (New York: Palgrave Macmillan).

Fagernäs, Sonja, and John Roberts, 2004. “Fiscal Impact of Aid: A Survey of Issues andSynthesis of Country Studies of Malawi, Uganda and Zambia,” ESAU WorkingPaper No. 11 (London: Overseas Development Institute, Economic and StatisticsAnalysis Unit).

Feeny, Simon, and Mark McGillivray, 2003, “Aid and Public Sector Borrowing inDeveloping Countries,” Journal of International Development, Vol. 15, pp. 989–98.

Fischer, Stanley, 1993, “The Role of Macroeconomic Factors in Growth,” Journal ofMonetary Economics, Vol. 32, No. 3, pp. 485–512.

Franco-Rodriguez, Susana, 2000, “Recent Developments in Fiscal Response with anApplication to Costa Rica,”Journal of International Development, Vol. 12, Part 3, pp. 429–42.

———, Oliver Morrissey, and Mark McGillivray, 1998, “Aid and the Public Sector inPakistan: Evidence with Endogenous Aid,” World Development, Vol. 26, No. 7, pp. 1241–50.



http://www.imf.org.uk/pppg/cape/Absorptive%20Capacity%20Seminar%20-%20Final%20Report.pdf

http://www.imf.org.uk/pppg/cape/Absorptive%20Capacity%20Seminar%20-%20Final%20Report.pdf

Gang, Ira N., and Haider A. Khan, 1990, “Foreign Aid, Taxes and Public Investment,”Journal of Development Economics, Vol. 34, Nos. 1–2, pp. 355–69.

Gatheru, Wamuyu, and Robert Shaw, eds., 1998, Our Problems, Our Solutions: AnEconomic and Public Policy Agenda for Kenya (Nairobi: Institute of Economic Affairs).

Ghosh, Atish, and Steven Phillips, 1998, “Warning: Inflation May Be Harmful to YourGrowth,” Staff Papers, International Monetary Fund, Vol. 45, pp. 672–710.

Ghura, Dhaneshwar, 2002, “Tax Revenue in Sub-Saharan Africa: Effects of EconomicPolicies and Corruption,” in Governance, Corruption, and Economic Performance, ed. byGeorge T. Abed and Sanjeev Gupta (Washington: International Monetary Fund).

———, and Goodwin, Barry, 2000. “Determinants of Private Investment: A Cross-Regional Empirical Investigation,” Applied Economics, Vol. 32, No. 14, pp. 1819–29.

———, and Thomas J. Grennes, 1993, “The Real Exchange Rate and MacroeconomicPerformance in Sub-Saharan Africa, Journal of Development Economics, Vol. 42, No. 1,pp. 155–74.

Greene, Joshua E., and Delano Villanueva, 1991, “Private Investment in DevelopingCountries: An Empirical Analysis,” Staff Papers, International Monetary Fund, Vol. 38, No. 1, pp. 33–58.

Group of Eight, 2005, Gleneagles Communiqué on Africa, Climate Change, Energy andSustainable Development (London: U.K. Foreign and Commonwealth Affairs Office).Available at: http://www.g8.gov.uk

Guillaumont, Patrick, and Lisa Chauvet, 2001, “Aid and Performance: AReassessment,” Journal of Development Studies, Vol. 37, No. 6, pp. 66–92.

Gupta, Kanhaya, 1993, “Sectoral Fungibility of Foreign Aid: Evidence from India”(unpublished; Edmonton, Alberta, Canada: University of Alberta).

Gupta, Sanjeev, Benedict Clements, Emanuele Baldacci, and Carlos Mulas-Granados,2004, “Fiscal Policy, Expenditure Composition, and Growth in Low-IncomeCountries,” in Helping Countries Develop: The Role of Fiscal Policy, ed. by SanjeevGupta, Gabriela Inchauste, and Benedict Clements (Washington: InternationalMonetary Fund, 2004), chapter 2.

Gupta, Sanjeev, Benedict Clements, Alexander Pivovarsky, and Erwin R. Tiongson,2004, “Foreign Aid and Revenue Response: Does the Composition of AidMatter?” in Helping Countries Develop: The Role of Fiscal Policy, ed. by Sanjeev Gupta,Gabriela Inchauste, and Benedict Clements (Washington: InternationalMonetary Fund).

Gupta, Sanjeev, Hamid R. Davoodi, and Rosa Alonso-Terme, 2002, “Does CorruptionAffect Income Inequality and Poverty?” in Governance, Corruption, and EconomicPerformance, ed. by George T. Abed and Sanjeev Gupta (Washington: InternationalMonetary Fund).

Gupta, Sanjeev, Hamid R. Davoodi, and Erwin R. Tiongson, 2002, “Corruption and theProvision of Health Care and Education Services,” in Governance, Corruption, andEconomic Performance, ed. by George T. Abed and Sanjeev Gupta (Washington:International Monetary Fund), chapter 10.

Gupta, Sanjeev, Luiz de Mello, and Raju Sharan, 2002,”Corruption and MilitarySpending,” in Governance, Corruption, and Economic Performance, ed. by George T.Abed and Sanjeev Gupta (Washington: International Monetary Fund).

Gylfason, Thorvaldur, and Tryggvi T. Herbertsson, 2001, “Does Inflation Matter forGrowth?” Japan and the World Economy, Vol. 13, pp. 405–28.

Haacker, Markus, 2004a, The Macroeconomics of HIV/AIDS (Washington: InternationalMonetary Fund).

References 63


http://www.g8.gov.uk

———, 2004b, “The Impact of HIV/AIDS on Government Finance and PublicServices,” in The Macroeconomics of HIV/AIDS, ed. by Markus Haacker (Washington:International Monetary Fund), chapter 7.

———, 2005, “Fiscal and Macroeconomic Aspects of the Response to HIV/AIDS,” inThe HIV Pandemic: Local and Global Implications, ed. by Eduard Beck and others(forthcoming; New York: Oxford University Press).

Hadjimichael, Michael T., and Dhaneshwar Ghura, 1995, “Public Policies and PrivateSavings and Investment in Sub-Saharan Africa: An Empirical Investigation,” IMFWorking Paper No. 95/19 (Washington: International Monetary Fund).

Hansen, Henrik, and Finn Tarp, 2001, “Aid and Growth Regressions,”Journal ofDevelopment Economics, Vol. 64, No. 2, pp. 547–70.

Hausmann, Ricardo, Lant Pritchett, and Dani Rodrik, 2004, “Growth Accelerations,”NBER Working Paper 10566 (Cambridge, Massachusetts: National Bureau ofEconomic Research).

Hausmann, Ricardo, and Roberto Rigobon, 2003. “An Alternative Interpretation of the‘Resource Curse:’ Theory and Policy Implications,” in Fiscal Policy Formulation andImplementation in Oil-Producing Countries, ed. by Jeffrey M. Davis, Rolando Ossowski,and Annalisa Fedelino (Washington: International Monetary Fund), chapter 2.

Heller, Peter S., 1975, “A Model of Public Fiscal Behavior in Developing Countries:Aid, Investment, and Taxation,”American Economic Review, Vol. 65, No. 3, pp. 429–45.

———, 2005, “Pity the Finance Minister: Issues in Managing a Substantial Scaling-Upof Aid Flows,” IMF Working Paper 05/180 (Washington: International MonetaryFund).

———, and Sanjeev Gupta, 2004, “More Aid: Making It Work for the Poor,” in HelpingCountries Develop: The Role of Fiscal Policy, ed. by Sanjeev Gupta, Gabriela Inchauste,and Benedict Clements (Washington: International Monetary Fund), chapter 15.

Hood, Ron, David Husband, and Fei Yu, 2002, “Recurrent Expenditure Requirementsof Capital Projects: Estimation for Budget Purposes,” Policy Research WorkingPaper No. 2938 (Washington: World Bank).

International Monetary Fund (IMF), 2004a, “Public Investment and Fiscal Policy”(unpublished; Washington: International Monetary Fund). Available at:http://www.imf.org/external/np/fad/2004/pifp/eng/pifp.pdf

———, 2004b, “Rebuilding Fiscal Institutions in Post-Conflict Countries” (unpub-lished; Washington: International Monetary Fund). Available at: http://www.imf.org/external/np/fad/2005/022505.htm

———, 2004c, “Public-Private Partnerships,” (unpublished; Washington InternationalMonetary Fund). Available at: http://www.imf.org/external/np/fad/2004/pifp/eng/031204.htm

———, 2005a, “The Federal Democratic Republic of Ethiopia: 2004 Article IVConsultation and Sixth Review Under the Three-Year Arrangement Under thePoverty Reduction and Growth Facility,” IMF Staff Country Report No. 05/25(Washington: International Monetary Fund).

———, 2005b, “The Federal Democratic Republic of Ethiopia: Debt SustainabilityAnalysis,” IMF Staff Country Report No. 05/27 (Washington: InternationalMonetary Fund).

———, 2005c, “The Federal Democratic Republic of Ethiopia: Selected Issues andStatistical Appendix,” IMF Staff Country Report No. 05/25 (Washington:International Monetary Fund).

———, 2005d, “The Macroeconomics of Managing Increased Aid Flows: Experiencesof Low-Income Countries and Policy Implications” (Washington: International



http://www.imf.org/external/np/fad/2004/pifp/eng/pifp.pdf

http://www.imf.org/external/np/fad/2005/022505.htm

http://www.imf.org/external/np/fad/2004/pifp/eng/031204.htm

http://www.imf.org/external/np/fad/2004/pifp/eng/031204.htm

http://www.imf.org/external/np/fad/2005/022505.htm

Monetary Fund). Available at: http://www.imf.org/external/np/pp/eng/2005/080805a.htm

———, 2005e, “Public Investment and Fiscal Policy—Summaries of the Pilot CountryStudies” (unpublished; Washington: International Monetary Fund). Available at:http://www.imf.org/external/np/pp/eng/2005/040105b.htm

———, 2005f, “Monetary and Fiscal Policy Design Issues in Low Income Countries”(unpublished; Washington: International Monetary Fund). Available at:http://www.imf.org/external/np/pp/eng/2005/080805m.htm.

———, and World Bank, 2004, “Debt Sustainability in Low-Income Countries: FurtherConsiderations on an Operational Framework and Policy Implications”(Washington: IMF and International Development Association). Available at:http://www.imf.org/external/np/pdr/sustain/2004/091004.pdf

Iqbal, Zafar, 1997, “Foreign Aid and the Public Sector: A Model of Fiscal Behaviour inPakistan,”Pakistan Development Review, Vol. 36, No. 2, pp. 115–29.

Islam, Muhammed N., 2003, “Political Regimes and the Effects of Foreign Aid onEconomic Growth,” Journal of Developing Areas, Vol. 37, No. 1, pp. 35–53.

Joint United Nations Program on HIV/AIDS (UNAIDS), 2004, 2004 Report on theGlobal AIDS Epidemic (Geneva: UNAIDS).

Kasekende, Louis A., and Michael Atingi-Ego, 1999, “Uganda’s Experience with Aid,”Journal of African Economies, Vol. 8, No. 4, pp. 617–49.

Keen, Michael, and Alejandro Simone, 2004, “Tax Policy in Developing Countries:Some Lessons from the 1990s and Some Challenges Ahead,” in Helping CountriesDevelop: The Role of Fiscal Policy, ed. by Sanjeev Gupta, Gabriela Inchauste, BenedictClements (Washington: International Monetary Fund), chapter 12.

Khan, Haider Ali, and Eiichi Hoshino, 1992, “Impact of Foreign Aid on the FiscalBehavior of LDC Governments,” World Development, Vol. 20, No. 10, pp. 1481–88.

Khan, Mohsin S., and Manmohan S. Kumar, 1997, “Public and Private Investment andthe Growth Process in Developing Countries,”Oxford Bulletin of Economics andStatistics, Vol. 59, No. 1, pp. 69–88.

Khan, Mohsin S., and Abdelhak S. Senhadji, 2001, “Threshold Effects in theRelationship between Inflation and Growth,” IMF Staff Papers, InternationalMonetary Fund, Vol. 48, pp. 1–21.

Khilji, Nasir M., and Ernest M. Zampelli, 1991, ‘The Fungibility of U.S. Assistance toDeveloping Countries and the Impact on Recipient Expenditures: A Case Study ofPakistan,” World Development, Vol. 19, No. 8, pp. 1095–105.

Klitgaard, Robert, R. Maclean-Abaroa, and H. Lindsey Parris, 1999, Corrupt Cities: APractical Guide to Cure and Prevention, (Oakland, California: ICS Press).

Kochhar, K. and Sharmini Coorey, 1999, “Economic Growth: What Has Been Achievedand How?” in Economic Adjustment and Reform in Low-Income Countries: Studies by the IMF Staff, ed. by H. Bredenkamp and S. Schadler (Washington: InternationalMonetary Fund), pp. 71–98.

Krueger, Alan B., and Mikael Lindahl, 2001,“Education for Growth: Why and forWhom?” Journal of Economic Literature, Vol. 39, No. 4, pp. 1101–36.

Lee, Jang-Yung, 1997, “Sterilizing Capital Inflows,” Economic Issues No. 7 (Washington:International Monetary Fund).

Leite, Carlos, and Jens Weidmann, 2002, “Does Mother Nature Corrupt? NaturalResources, Corruption, and Economic Growth,” in Governance, Corruption, andEconomic Performance, ed. by George T. Abed and Sanjeev Gupta (Washington:International Monetary Fund), chapter 7.

Leuthold, Jane H., 1991, “Tax Shares in Developing Economies: A Panel Study,”Journalof Development Economics, Vol. 35, No. 1, pp. 173–85.

References 65


http://www.imf.org/external/np/pp/eng/2005/080805a.htm

http://www.imf.org/external/np/pp/eng/2005/040105b.htm

http://www.imf.org/external/np/pp/eng/2005/080805m.htm

http://www.imf.org/external/np/pdr/sustain/2004/091004.pdf

http://www.imf.org/external/np/pp/eng/2005/080805a.htm

Lofgren, Hans, and Carolina Diaz-Bonilla, 2005, “An Ethiopian Strategy for Achievingthe Millennium Development Goals: Simulations with the MAMS Model” (unpub-lished; Washington: World Bank).

Masha, Iyabo, 2004, “An Economic Assessment of Botswana’s National StrategicAssessment for HIV/AIDS,” in The Macroeconomics of HIV/AIDS, ed. by MarkusHaacker (Washington: International Monetary Fund), chapter 9.

Mauro, Paolo, 1995, “Corruption and Growth,”Quarterly Journal of Economics, Vol. 110,No. 3, pp. 681–712.

———, 1998, “Corruption and the Composition of Government Expenditure, Journalof Public Economics, Vol. 69, No. 2, pp. 263–79. Reprinted in Governance, Corruption,and Economic Performance, ed. by George T. Abed and Sanjeev Gupta (Washington:International Monetary Fund, 2002), chapter 9.

McGillivray, Mark, 2000, “Aid and Public Sector Behavior in Developing Countries,”Review of Development Economics, Vol. 4, No. 2, pp. 156–63.

———, and Akhter Ahmed, 1999, “Aid, Adjustment and Public Sector Fiscal Behaviourin the Philippines,”Journal of the Asia Pacific Economy, Vol. 4, No. 2, pp. 381–91.

McGillivray, Mark, and Oliver Morrissey, 2001, “A Review of Evidence on the FiscalEffects of Aid,” CREDIT Research Paper No. 01/13 (Nottingham, England: Centrefor Research in Economic Development and International Trade, University ofNottingham).

Mosley, Paul, 1980, “Aid, Savings and Growth Revisited,”Oxford Bulletin of Economics andStatistics, Vol. 42, No. 2, pp. 79–95.

Nkusu, Mwanza, 2004, “Aid and the Dutch Disease in Low-Income Countries:Informed Diagnosis for Prudent Prognoses,” IMF Working Paper 04/49(Washington: International Monetary Fund).

Nyoni, Timothy S., 1998, “Foreign Aid and Economic Performance in Tanzania,”World Development, Vol. 26, No. 7, pp. 1235–40.

Odedokun, Matthew O., 1997, “Relative Effects of Public Versus Private InvestmentSpending on Economic Efficiency and Growth in Developing Countries,” AppliedEconomics, Vol. 29, No. 10, pp. 1325–36.

Otim, Samuel, 1996, “Foreign Aid and Government Fiscal Behaviour in Low-IncomeSouth Asian Countries,” Applied Economics, Vol. 28, No. 8, pp. 927–33.

Pack, Howard, and Janet R. Pack,1990, “Is Foreign Aid Fungible? The Case ofIndonesia,” Economic Journal, Vol. 100, No. 399, pp. 188–94.

———, 1993, “Foreign Aid and the Question of Fungibility,”Review of Economics andStatistics, Vol. 75, No. 2, pp. 258–65.

Prati, Alessandro, Ratna Sahay, and Thierry Tressel, 2003, “Is There a Case forSterilizing Foreign Aid,” Macroeconomic Challenges in Low-Income CountriesWorkshop (Washington: International Monetary Fund). Available at:http://www.imf.org/external/np/res/seminars/2003/lic/pdf/tt.pdf

Prati, Alessandro, and Thierry Tressel, 2005, “What Monetary Policy for Aid-ReceivingCountries?” presented at the United Nations DESA Development Forum, UnitedNations, March 14–15 (forthcoming in conference volume; New York: UnitedNations).

Radelet, Steven, 2004, “Aid Effectiveness and the Millennium Development Goals,”Working Paper No 39 (Washington: Center for Global Development).

Rajan, Raghuram G., and Arvind Subramanian, 2005a, “What Undermines Aid’sImpact on Growth?” IMF Working Paper 05/126 (Washington: InternationalMonetary Fund).



http://www.imf.org/external/np/res/seminars/2003/lic/pdf/tt.pdf

———, 2005b, “Aid and Growth: What Does the Cross-Country Evidence ReallyShow?” IMF Working Paper 05/127 (Washington: International Monetary Fund).

Rawlings, Laura, Lynne Sherburne-Benz, and Julie Van Domelen, 2004, EvaluatingSocial Fund Performance: A Cross-Country Analysis of Community Investments(Washington: World Bank).

Razin, Ofair, and Susan M. Collins, 1997, “Real Exchange Rate Misalignments andGrowth,” NBER Working Paper No. 6174 (Cambridge, Massachusetts: NationalBureau of Economic Research).

Romp, Ward, and Jakob De Haan, 2005, “Public Capital and Economic Growth: ACritical Survey,” European Investment Bank Papers, Vol. 10, No. 1.

Rubino, C., 1997, “Aid, The Public Sector and the Real Exchange Rate: The Case ofIndonesia” (Unpublished Ph.d. thesis; Warwick, United Kingdom: Department ofEconomics, University of Warwick).

Sackey, Harry A., 2002, “External Aid Flows and the Real Exchange Rate in Ghana,”AERC Research Paper 110 (Nairobi: African Economic Research Consortium).

Sarel, Michael, 1996, “Nonlinear Effects of Inflation on Economic Growth,” StaffPapers, International Monetary Fund, Vol. 43, pp. 199–215.

Sekkat, Khalid, and Aristomene Varoudakis, 2000, “Exchange Rate Management andManufactured Exports in Sub-Saharan Africa,” Journal of Development Economics,Vol. 61, No. 1, pp. 237–53.

Singh, Ram D., 1985, “State Intervention, Foreign Economic Aid, Savings and Growthin LDCs: Some Recent Evidence,” Kyklos, Vol. 38, No. 2, pp. 216–32.

Summers, Todd, and Jennifer Kates, 2003, Global Funding for HIV/AIDS in Resource-PoorSettings (Washington: Kaiser Family Foundation).

Sundberg, Mark, Hans Lofgren, and François Bourguignon, 2005, “AbsorptiveCapacity and Achieving the MDGs: The Case of Ethiopia,” DevelopmentEconomics Working Paper (Washington: World Bank).

Swaroop, Vinaya, Shikha Jha, and Andrew Sunil Rajkumar 2000, “Fiscal Effects ofForeign Aid in a Federal System of Governance: The Case of India,” Journal ofPublic Economics, Vol. 77, No. 3, pp. 307–30.

Tanzi, Vito, and Hamid R. Davoodi, 1998, “Corruption, Public Investment, andGrowth,” in The Welfare State, Public Investment, and Growth, ed. by Hirofumi Shibataand Toshihiro Ihori (Tokyo: Springer Verlag).

———, 2002, “Corruption, Growth, and Public Finances,” in Governance, Corruption,and Economic Performance, ed. by George T. Abed and Sanjeev Gupta (Washington:International Monetary Fund), chapter 8.

Transparency International, 2001, Global Corruption Report 2001 (Berlin).

Transparency International Kenya, 2002, Kenya Urban Bribery Index 2002 (Nairobi).

U.K. Department for International Development (DfID), 2004, Macroeconomic Issues forScaling Up Aid Flows, “How to Note” (London).

UN Millennium Project, 2005. Investing in Development: A Practical Plan to Achieve theMillennium Development Goals. (New York: United Nations Development Program).

Van Rijckeghem, Caroline, and Beatrice Weder, 2002, “Bureaucratic Corruption andthe Rate of Temptation: Do Wages in the Civil Service Affect Corruption and byHow Much?” in Governance, Corruption, and Economic Performance, ed. by George T.Abed and Sanjeev Gupta (Washington: International Monetary Fund).

van Wijnbergen, Sweder, 1985, “Aid, Export Promotion, and the Real Exchange Rate: AnAfrican Dilemma?” CEPR Discussion Paper No. 88 (London: Centre for EconomicPolicy Research).

References 67


World Bank, 2004, World Development Report 2004 (Washington).

———, 2005, Global Development Finance 2005: Mobilizing Finance and ManagingVulnerability (Washington).

World Bank and IMF, 2005, Global Monitoring Report 2005 (Washington).

Yang, Yongzheng, and Sanjeev Gupta, 2005, “Regional Trade Arrangements in Africa:Past Performance and the Way Forward,” IMF Working Paper 05/36 (Washington:International Monetary Fund).

Yano, Makoto, and Jeffrey B. Nugent, 1999, “Aid, Nontraded Goods, and the TransferParadox in Small Countries,” American Economic Review, Vol. 89, No. 3, pp. 431–49.

Younger, Stephen D., 1992, “Aid and the Dutch Disease: Macroeconomic ManagementWhen Everybody Loves You,” World Development, Vol. 20, No. 11, pp. 1587–97.



macroeconomic challenges of scaling up aid to africa...millennium project (2005) has argued for $33...

Documents